1 2 3 4 5 the Atypical RNA-Binding Protein TAF15 Regulates Dorsoanterior Neural Development 6 Through Diverse Mechanisms in Xeno

bioRxiv preprint doi: https://doi.org/10.1101/2021.06.14.041913; this version posted June 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

1 2 3 4 5 6 The atypical RNA-binding protein TAF15 regulates dorsoanterior neural development 7 through diverse mechanisms in Xenopus tropicalis 8 9 10 11 12 13 14 15 Caitlin S. DeJong 1*, Darwin S. Dichmann 1#, Cameron R. T. Exner 2, Yuxiao Xu 2, & 16 Richard M. Harland 1‡ 17 18 19 20 21 22 1 Molecular and Cell Biology Department, Genetics, Genomics and Development 23 Division, University of California, Berkeley, CA, USA 24 2 Department of Psychiatry, Weill Institute for Neurosciences, Quantitative Biosciences 25 Institute, University of California San Francisco, San Francisco, CA 26 * Present address: Vaccine and Infectious Disease Division, Fred Hutchinson Cancer 27 Research Center, Seattle, Washington 98109; # Present address: Invitae Corporation, 28 San Francisco, California, USA 29 30 31 32 33 34 35

37 ‡To whom correspondence should be addressed. Mail: [email protected] Running Title: TAF15 regulates dorsoanterior neural development in Xenopus tropicalis Page 1

38 ABSTRACT 39 40 The FET family of atypical RNA-binding proteins includes Fused in sarcoma (Fus), 41 Ewing’s sarcoma (EWS), and the TATA-binding protein-associate factor 15 (TAF15). 42 FET proteins are highly conserved, suggesting specialized requirements for each 43 protein. Fus regulates splicing of transcripts required for mesoderm differentiation and 44 cell adhesion in Xenopus, but roles that EWS and TAF15 play remain unknown. Here 45 we analyze the roles of maternally deposited and zygotically transcribed TAF15, which 46 is essential for the proper development of dorsoanterior neural tissues. By measuring 47 changes in exon usage and transcript abundance from TAF15-depleted embryos we 48 found TAF15 may regulate dorsoanterior neural development through fgfr4 and 49 ventx2.1. TAF15 uses distinct mechanisms to downregulate FGFR4 expression: 1) 50 retention of a single intron within fgfr4 when maternal and zygotic TAF15 is depleted, 51 and 2) reduction of total fgfr4 transcript when zygotic TAF15 alone is depleted. The two 52 mechanisms of gene regulation (post-transcriptional vs transcriptional) suggest TAF15- 53 mediated gene regulation is target and cofactor-dependent, depending on the milieu of 54 factors that are present at different times of development. 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85

Running Title: TAF15 regulates dorsoanterior neural development in Xenopus tropicalis Page 2

86 INTRODUCTION 87 88 The FET family of atypical RNA-binding proteins includes Fused in sarcoma (Fus), 89 Ewing’s sarcoma (EWS), and the TATA-binding protein-associate factor 15 (TAF15). 90 This is a family of heterogeneous nuclear ribonuclear particle (hnRNP) proteins that 91 contain domains for transcriptional activation, RNA binding, and DNA binding 92 (Schwartz, Cech, & Parker, 2015). FET family proteins function in both RNA 93 Polymerase II-mediated transcription and pre-mRNA splicing (Schwartz et al., 2015; 94 Tan & Manley, 2009). Among vertebrates, the three FET members are highly conserved 95 from fish to mammals, suggesting an independent and specialized requirement for each 96 protein (Schwartz et al., 2015). FET proteins have been investigated primarily as 97 components of fusion oncogenes; following abnormal chromosomal translocations, FET 98 protein N-terminal low-complexity/activation domains are found fused to various DNA- 99 binding proteins, contributing to the formation of various cancers (e.g. sarcomas and 100 leukemias) as well as neuronal degenerative diseases (Crozat, Åman, Mandahl, & Ron, 101 1993; Delattre et al., 1992; King, Gitler, & Shorter, 2012; Kovar, 2011; Martini et al., 102 2002; Neumann et al., 2011; Panagopoulos et al., 1999; Rabbitts, Forster, Larson, & 103 Nathan, 1993; Sjögren, Meis-Kindblom, Kindblom, Åman, & Stenman, 1999; Tan & 104 Manley, 2009; Vance et al., 2009). It has only been more recently that the functions of 105 these proteins have been examined in their full length, “wild-type”, form (Dichmann & 106 Harland, 2012; Schwartz et al., 2015; Tan & Manley, 2009). Studies of the structural, 107 functional, and biochemical properties of the FET family proteins determined that these 108 proteins have multiple functions, such that FET proteins may have evolved to facilitate 109 the complex coupling of transcription and mRNA processing that occurs in multicellular 110 organisms (Kato et al., 2012; Schwartz et al., 2015; Schwartz, Wang, Podell, & Cech, 111 2013). 112 113 The majority of work that has contributed to our understanding of FET protein biology 114 and disease mechanism has been carried out in cell lines and mouse models (Hicks et 115 al., 2000; Li et al., 2007; Scekic‐Zahirovic et al., 2016; Sharma et al., 2016; Svetoni et 116 al., 2016; Kapeli et al., 2016) with little known of the role of FET proteins in embryonic 117 development. Previous work from our lab examining the role of Fus in Xenopus 118 development found that embryos depleted of Fus exhibit mesoderm differentiation 119 defects and epithelial dissociation (Dichmann & Harland, 2012). The underlying 120 mechanism of these phenotypes was retention of all introns in fibroblast growth factor 8 121 (fgf8), fibroblast growth factor receptor 2 (fgfr2), and cadherin 1 (cdh1) transcripts 122 (Dichmann & Harland, 2012). This study therefore showed that Fus is required for 123 processing of a subset of transcripts in Xenopus development. Given the important role 124 of FUS in Xenopus development, the perplexing potential for functional redundancy of 125 FET family members in mouse (while remaining highly conserved throughout 126 vertebrates), and the lack of basic research on FET protein functions, we examined the 127 role of TAF15 in early Xenopus development; including the role of maternal versus 128 zygotic TAF15. 129 130 To determine the role of TAF15 in early Xenopus development, we used RNA- 131 sequencing (RNAseq) from single embryos depleted of maternal (M) and zygotic (Z) 132 TAF15, using reagents that target all mRNA by inhibiting translation with Morpholino 133 oligonucleotides, or just zygotic function using splice blocking MOs or CRISPR

Running Title: TAF15 regulates dorsoanterior neural development in Xenopus tropicalis Page 3

134 mediated mutagenesis. Upon evaluating the transcriptional changes that result from 135 M+Z versus Z-only TAF15 depletion, we find a subset of target genes whose expression 136 is regulated either post-transcriptionally (via intron retention) or by transcript level, 137 depending on whether maternal or zygotic TAF15 is depleted. These results suggest 138 that maternal TAF15 translation is limiting for splicing of a subset of mRNAS. Further, 139 we show that during the time of zygotic genome activation, zygotic TAF15 modulates 140 the expression of nascent target genes, acting at the transcriptional (rather than post- 141 transcriptional) level. Interestingly, we find that in at least one case that we examined 142 closely, maternal and zygotic TAF15 have a shared target gene (fgfr4), but that each act 143 to regulate the target gene expression through post-transcriptional vs. transcriptional 144 mechanisms, respectively. Here, we describe our findings as an example in Xenopus 145 where the gene product TAF15: 1) uses distinct molecular mechanisms to regulate the 146 expression of the same gene target (fgfr4) depending on the time of development in 147 which TAF15 is expressed (maternal versus zygotic) and 2) ensures proper 148 dorsoanterior neural development through two distinct molecular pathways (fgfr4 and 149 ventx2.1). 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181

Running Title: TAF15 regulates dorsoanterior neural development in Xenopus tropicalis Page 4

182 METHODS 183 184 Ethics statement 185 This study was carried out in strict accordance with the recommendations in the Guide 186 for the Care and Use of Laboratory Animals of the National Institutes of Health. The 187 protocol was approved by the Animal Care and Use Committee at the University of 188 California, Berkeley. 189 190 General Xenopus Embryo Culture 191 Xenopus tropicalis embryos were obtained through natural matings. For next day 192 (daytime) matings, males were housed individually, and females were housed together, 193 in four liter Rubbermaid® containers filled with two liters of water collected from the X. 194 tropicalis housing racks. The night before the natural mating, males were boosted with 195 100 units (U) of human chorionic gonadotropin (HCG: Chorulon®, Merck, NADA 196 NO.140-927, Code No. 133754) and females were primed with 10U HCG. The morning 197 of mating, females were boosted with 200U HCG and paired with males. X. tropicalis 198 embryos were collected using a disposable polyethylene transfer pipet (Fisherbrand®, 199 Cat No. 13-711-7M), with the tip cut off to enlarge the opening. Embryos were dejellied 200 and cultured as previously described (Khokha et al., 2002). 201 202 Xenopus tropicalis embryos were allowed to develop in 1/9X Marc’s Modified Ringer 203 (MMR), until desired stage according to the normal table of development (Nieuwkoop & 204 Faber, 1994). 205 206 Whole-mount RNA in situ Hybridization 207 Xenopus embryos were fixed in MEMFA (0.1M MOPS pH7.4, 2mM EGTA, 1mM 208 MgSO4, 3.7% v/v Formaldehyde) as previously described (Sive et al., 2000). RNA 209 probes were labelled with digoxigenin-UTP, and chromogenic reactions were carried out 210 by incubating hybridized embryos in Anti-Digoxigenin-AP Fab fragments, 1:3000 211 (Roche, 11 093 274 910), and the alkaline phosphatase substrate BM purpled (Roche, 212 11 442 074 001), as previously described (Sive et al., 2000). Xenopus tropicalis 213 embryos were incubated in prehybridization buffer for at least one hour. 214 215 Western Blotting 216 Xenopus tropicalis embryos were lysed in 20mM Tris–HCl pH 8.0, 50mM NaCl, 2mM 217 EDTA, with 1x protease inhibitor (Roche cOmplete, Mini, EDTA-free, 218 Product#11836170001), with 20uL per embryo 1% Triton™ X 100 detergent (freshly 219 added ;Sigma T8787), homogenized by pipetting and freezing at -80°C. To pellet debris, 220 lysates were spun at 2655 RCF (eppendorf Centrifuge 5417C) at 4°, and supernatant 221 was transferred to new tube. To clear embryos of yolk, lysates were spun at 2655 RCF 222 an additional two to three times, each time using a vacuum with a non-filtered p200 tip 223 to briskly remove the yolk from top of the lysate. Lysate protein concentrations were 224 measured with Bradford assays (Bio-Rad Protein Assay Dye Reagent Concentrate, 225 Cat#500-0006) and read using a Molecular Devices, Spectramax M2 plate reader. 226 Lysates were aliquoted for use and 6x loading dye was added. Samples were heated at 227 80°C for 10 minutes. Lysates were run on 8% polyacrylamide gel and run at 120V for 1 228 hour and forty-five minutes, eliminating proteins of 10-20 kDa. Proteins were transferred 229 from gels using the semi-dry transfer system (BioRad Trans-Blot® SD Semi-Dry Running Title: TAF15 regulates dorsoanterior neural development in Xenopus tropicalis Page 5

230 Electrophoretic Transfer Cell #170-3940) to Immobilon®-FL transfer membranes, PVDF 231 (Millipore, IPFL00010), and blocked for one hour at room temperature with 1X 232 Odyssey® Blocking Buffer (PBS) (LI-COR, 927- 40000). Anti-TAF15 (TAFII68) antibody 233 (Bethyl Laboratories, A300-309A) was used at 1:3000, anti-FGFR4 (CD334) antibody 234 (Thermo Scientific, PA5-28175) was used at 1:2000, and anti-β-actin antibody 235 (GeneTex, clone GT5512, GTX629630) was used at 1:5000, diluted in 5% BSA in TBS- 236 tween (TBS-T), and incubated overnight at 4°C. Fluorescent secondary antibodies, 237 Alexa Fluor® 680 goat anti- Rabbit IgG (Invitrogen, A-21109), IRDye® 800CW Donkey 238 anti-Mouse IgG (LI-COR, 925-32212), were incubated at 1:10,000 for one hour at room 239 temperature in the dark. Western blots were visualized and quantified using a LI-COR 240 imager and software (LI-COR, Odyssey). 241 242 Microinjection of Morpholino Antisense Oligonucleotides: Maternal and/or 243 Zygotic TAF15 Depletion 244 Morpholino antisense and mismatch oligos (MOs) were designed and ordered from 245 GeneTools LLC. 246 taf15 translation-blocking (Maternal and Zygotic) MO: 247 5’-AGCTACTGGGATCTGAAGACATGAT-3’; 248 taf15 splice-blocking (Zygotic only) MO: 5’-TTCCAAAACCTACCTTTGTTGCTGC-3’; 249 Mismatch MO: ‘5-AGCTAGTCGCATCTCAACACATGAT-3’. 250 251 MOs were dissolved in nuclease-free water to 8.5ng/nL (1mM). Translation-blocking 252 (17ng/cell), splice-blocking (8ng/cell), or mismatch (17ng/cell) MOs were injected into 253 either one of two, or two of two cells, to deplete target mRNA from half or the whole 254 embryo, respectively. To trace which cells contain MO, each MO was coinjected with 255 the fluorescein-conjugated standard control oligo (GeneTools). 256 257 RNA-extraction 258 RNA from single Xenopus tropicalis embryos was isolated using the Trizol® Reagent 259 (Ambion Ref# 15596026), optimized for extracting small amounts of RNA from single 260 Xenopus tropicalis embryos. Single embryos were collected in 200μl of Trizol® Reagent 261 (Ambion Ref# 15596026), homogenized by pipetting, and stored at -80°C for a minimum 262 of fourteen hours. Homogenized samples were thawed and incubate at room 263 temperature (RT) for five minutes to allow complete dissociation of the nucleoprotein 264 complex. 40uL of chloroform was added to each sample, vortexed, and incubated at RT 265 for 2-3 minutes. Samples were centrifuged at 12,000 RCF for 15 minutes at 4°C and 266 90uL of the upper aqueous phase was removed and placed in new tube. RNA was 267 precipitated with 100% isopropanol with 5ug of linear acrylamide (Ambion, AM9520) for 268 pellet detection. RNA pellets were washed with 75% ethanol and air dried at RT for 5-10 269 minutes. RNA pellets were resuspended in 250uL of Milli-Q water and gently pipetted 270 up and down and vortex. A second RNA extraction was performed by adding 250uL of 271 Acid-Phenol:Chloroform, pH 4.5 (with IAA, 125:24:1) (Ambion, Cat#AM9720) following 272 the manufacturer’s protocol. RNA was precipitated by adding 20uL 5M NH4OAC 273 (Ammonium Acetate) (Ambion AM9070G) and 220uL of 100% isopropanol and RNA 274 washed two times with 75% ethanol. RNA concentrations were measured using a 275 nanodrop (Nanodrop ND-1000 Spectrophotometer) and used for RNAseq library 276 preparation or qRT-PCR. 277

Running Title: TAF15 regulates dorsoanterior neural development in Xenopus tropicalis Page 6

278 RNA sequencing Library preparation and Analysis 279 Each paired-end library for RNA-sequencing (RNA-seq) was prepared using RNA 280 extracted from single Xenopus tropicalis embryos; each condition is composed of three 281 to four independently sequenced embryos. RNA-seq libraries were made strictly 282 following the Low Sample (LS) Protocol from TruSeq RNA Sample Preparation v2 283 Guide. The important modification made to this protocol was that all reagents were used 284 at half volume, except for the Bead Washing Buffer. 285 286 Final RNA-seq libraries were quantified using the KAPA Library Quantification Kits for 287 Illumina sequencing platforms (Roche, Kit code KK4824). 100bp paired-end sequencing 288 reads (Illumina HiSeq2000) were aligned to the Xenopus tropicalis genome version 7.1, 289 and an annotation from Darwin Dichmann (not published). RNA-seq data analysis for 290 differential gene expression was performed using both the Tuxedo Suite (Tophat, 291 Bowtie, Cufflinks, Cuffdiff) (Trapnell et al., 2012) and the Bioconductor package DESeq 292 (Love, Huber, & Anders, 2014). RNA-seq data analysis for differential intron-exon usage 293 was performed using the Bioconductor package, DEXseq (Reyes, Anders, & Huber, 294 2014) (www.bioconductor.org). In brief, read counts were normalized to library size per 295 feature. The p-values are the result of the statistical modeling peformed by 296 DESeq2/DEXSeq and indicates to which degree the difference in expression of a given 297 gene or exon is significant (in morphants compared to controls). The p-values were then 298 adjusted for false discovery rates (also performed by DESeq2/DEXSeq) for the number 299 of features since false positive calls are problematic when sampling a large number of 300 features (genes or exons) as is done in gene expression studies. 301 302 RNA-seq alignments were visualized using the Integrative Genomics Viewer (IGV) from 303 the Broad Institute (www.broadinstitute.org/igv/). 304 305 Complementary DNA (cDNA) Synthesis and qRT-PCR 306 RNA was isolated from single Xenopus tropicalis embryos as described above. cDNA 307 was synthesized using the iScript reverse transcription (RT) reaction protocol (BIO-RAD 308 Cat#170-8841). Optimally, 1ug of total RNA was reverse transcribed, but total RNA 309 yield from a single Xenopus tropicalis embryo varied. Therefore, all samples were 310 normalized to ensure all RT samples contained the same volume and concentration of 311 RNA. For each RT reaction, 7.5uL of 5x iScript RT Supermix was added to 30uL of 312 RNA/H2O. For no RT (NRT) samples, RNA samples with yields below a usable limit 313 were pooled and the sample volume was brought up to 30uL, and 7.5uL of 5x iScript 314 Supermix no RT was added. 315 316 cDNA samples were diluted to a working concentration of 5ng/uL and qRT-PCR 317 reactions were performed with the SsoAdvanced SYBR Green Supermix (BIO-RAD 318 Cat#172-5261) using a CFX96 Thermal Cycler (BIO-RAD) following the manufacturer’s 319 suggested protocol. NRT and no template control (NTC) samples were included for 320 each gene target. 321 322 Primer set pairs used in these studies: 323 324 eef1a1: F 5’-CCCTGCTGGAAGCTCTTGAC-3’; 325 R 5’-GGACACCAGTCTCCACACGA-3’

Running Title: TAF15 regulates dorsoanterior neural development in Xenopus tropicalis Page 7

326 fgfr4 intron 1: F 5’-AGGGCTAAGCAGTGCCTGTA-3’; 327 R 5’-AATGCAAGAGCAGCTCCAAT-3’ 328 fgfr4 total: F 5’-AGCCAGGAATGTTCTTGTGG-3’; 329 R 5’-TCCCATGTCAACACTCCAAA-3’ 330 ventx2.1: F 5’-AACAGCCAGCTGTCTCCAGT-3’; 331 R 5’-GCTGTGTCCCTGTGTAGCAA-3’ 332 engrailed 2: F 5’-GAAGACGACGACGACTGTCA-3’; 333 R 5’- AACTTTGCCTCCTCTGCTCA-3’ 334 bmp7.1 intron 3: F 5’- TCCCCTCCTCTATGGCTTTT-3’; 335 R 5’- AGTGGTGCCCAAGATTCAAC-3’ 336 bmp7.1 total: F 5’- CGGGAAAGGTTTGAAAATGA-3’; 337 R 5’- ATATCGAACACCAGCCATCC-3’ 338 cpl1 5’UTR: F 5’-TATGCCTCCCCTGTGGATAG-3’; 339 R 5’-TACTTTGCCTGCCCTATGCT-3’ 340 cpl1 total: F 5’-AGAGATCTGCCGCACTTTGT-3’; 341 R 5’-TGAACCACGTGGACCATAGA-3’ 342 dgka intron 9: F 5’-GAGGGTGATGTGACCATGTG-3’; 343 R 5’-TCCATTTTAAGCCCAACAGC-3’ 344 dgka intron 11: F 5’-GCGACCAATGAATGCTACCT-3’; 345 R 5’-TAAACATGCTGCTGGGTCAG-3’ 346 dgka total: F 5’-TGGATGAGAGGTGGATGTGA-3’; 347 R 5’-CATCACACAGTGGGGATGAG-3’ 348 pdgfa intron 1: F 5’-CCTCAGAGGCACTTTCCAAG-3’; 349 R 5’-TTGTGCTACAGAACCGCAAC-3’ 350 pdgfa total: F 5’-CCAGAGAAGCGTTCTGTTCC-3’; 351 R 5’-ACACACGGAGGCCAGATTAG-3’ 352 per2 intron 1: F 5’-GGCCTGTAATTTGGAGCTTTC-3’; 353 R 5’-CAAACCGGATGTGGGATTAT-3’ 354 per2 total: F 5’-TGGACGGGAATCAAGAAAAG-3’; 355 R 5’-GAACCCTCTCAGCAAACAGC-3’ 356 rab15 intron 2: F 5’-GGATGCTTTTGGTGGTGTTT-3’; 357 R 5’-ATTGGATGGTTTTGCCTCTG-3’ 358 rab15 total: F 5’-GGATGAAGCTTGCAGAGGAG-3’; 359 R 5’-CTCCAGCTCCCTTTTGTGAG-3’ 360 srsf4 intron 5: F 5’-TCTGATCCCCATTCAGTTGC-3’; 361 R 5’-GTTTTGCCTGCATAGCCAGT-3’ 362 srsf4 total: F 5’-GAGCAAGGATAGGGACCACA-3’; 363 R 5’-TCTCTTTGCTACGGCTGGAT-3’ 364 zdhhc5 intron 6: F 5’-GAAGCGGGCTATCTGAGTTG-3’; 365 R 5’-CCTTGGGGTCTGAGATTTGA-3’ 366 zdhhc5 total: F 5’-ACTGGCCAATTTTAGCATGG-3’; 367 R 5’-AACCTGCACTCCCCTTACCT-3’ 368 gpr110 intron 9: F 5’-GCCATTTAGTGATGGGCTGT-3’; 369 R 5’-CCAAGGCACACATACATGCT-3’ 370 gpr110 total: F 5’-GCTTGTCATATCCCGTTGCT-3’; 371 R 5’-CTAGCCCACACGTAGCCATT-3’ 372 isl1 intron 3: F 5’-CCCTCCTACCCTTTTCCAAG-3’; 373 R 5’-GGCTGAAGTGCGGAATCTAA-3’

Running Title: TAF15 regulates dorsoanterior neural development in Xenopus tropicalis Page 8

374 isl1 total: F 5’-ACTTTGCCCTGCAGAGTGAC-3’; 375 R 5’-TGCCTCTATAGGGCTGGCTA-3’ 376 377 Prior to use for quantitating transcripts, primer pair efficiency was tested on six standard 378 controls (serial dilutions of 1/5): 25ng = 2.50E+04, 5ng = 5.00E+03, 1ng = 1.00E+03, 379 0.2ng = 2.00E+02, 0.04ng = 4.00E+01, 0.008ng = 8.00E+00. 380 381 Glutaraldehyde Vibratome Sections 382 Following RNA in situ hybridization, embryos were selected for sectioning. Sectioning 383 was performed as previously described (Young et al., 2014). Embryos were equilibrated 384 in a PBS solution containing 20% sucrose, 30% BSA, 4.9% gelatin, and fixed with 1.5% 385 glutaraldehyde. Embryos were mounted into a Peel-A-Way® disposable embedding 386 mold (Polysciences, 18985). Once cured, the blocks were removed from the embedding 387 molds and cut into sectioning prisms. Prisms were affixed to Heroscape dice (Hasbro) 388 with super glue and mounted on a Pelco 101 Vibratome Series 1000 Sectioning System 389 (Ted Pella, Inc.), sections were made with razor blades (Personna). 50-75micron 390 sections were mounted on glass slides for imaging. 391 392 Microinjection of CRISPR/cas9: Zygotic taf15 Depletion 393 X.tropicalis taf15 sgRNA was designed using the Giraldez Lab CRISPRscan 394 (crisprscan.org). Sequence of X.tropicalis taf15 sgRNA (including PAM): 395 GCTATGGTGGTTATGGAGGAGG. sgRNA was generated by PCR amplification using 396 the Phusion Polymerase (NEB, M0530S) following the manufacturer’s protocol. 397 Forward primer 398 CTAGCtaatacgactcactataGGCTATGGTGGTTATGGAGGgttttagagctagaa and reverse 399 primer 400 AAAAGCACCGACTCGGTGCCACTTTTTCAAGTTGATAACGGACTAGCCTTATTTTAA 401 CTTGCTATTTCTAGCTCTAAAC were used for sgRNA amplification. RNA in vitro 402 transcription of sgRNA was performed using the T7 MEGAshortscript kit (Ambion/Life 403 Technologies, AM1354) and sgRNA cleanup was performed using the MEGAclear kit 404 (Ambion/Life Technologies, AM1908). 405 406 sgRNA injection cocktails included 1.5ng Cas9 protein, 400ng sgRNA, and Dextran555 407 (Life Technologies/Molecular Probes, D34679) diluted in water. Injection volume was 408 2nL. Embryos were injected in the two-cell embryo stage, injections were made into 409 either one of two, or two of two cells, to deplete target mRNA from half or the whole 410 embryo, respectively. 411 412 413 414 415 416 417 418 419 420 Running Title: TAF15 regulates dorsoanterior neural development in Xenopus tropicalis Page 9

421 RESULTS 422 423 taf15/TAF15 is both maternally deposited, and zygotically transcribed, with a 424 localized expression pattern.

425 To study of the role of TAF15 in Xenopus development, we first determined the time 426 and place of gene expression; both at the transcript and protein level (Figure 1). From 427 the RNA in situ hybridization (ISH) data (Figure 1A), we observe that taf15 is enriched in 428 the animal hemisphere through cleavage and gastrula stages (Figure 1A). During 429 neurulation, taf15 changes from diffuse dorsal expression (Figure 1A, stage 13), to 430 more specific expression around the neural plate but still throughout the ectoderm 431 (Figure 1A, stage 15). Enrichment in the tailbud stage (Figure 1A, stage 26), is in 432 dorsoanterior tissues of the embryo, particularly the brain, branchial arches, anlage of 433 the ear and eye, and pronephros. Lastly, in the early tadpole (Figure 1A, stage 32), 434 taf15 is enriched in the central nervous system (anterior and posterior), branchial 435 arches, otic vesicle, and the posterior domain of the eye. These data are consistent with 436 protein abundance in Western blots: TAF15 is deposited maternally and present 437 throughout embryogenesis, increasing in expression following zygotic genome 438 activation (ZGA) (Figure 1B).

439 taf15/TAF15 depletion leads to gross morphological defects.

440 To elucidate the role of TAF15 in development, we depleted both maternal and zygotic 441 (M+Z) TAF15 protein expression using a translation-blocking morpholino antisense 442 oligonucleotide (MO) and zygotic (Z) expression alone using either a splice-blocking MO 443 or CRISPR-associated protein-9 (Cas9) nuclease technology (CRISPR/Cas9) (Figure 444 1C-F). Following MO-mediated M+Z TAF15 depletion (by translation-blocking 445 morpholino), embryos exhibit gross morphological defects including a shortened 446 anterior-posterior axis, loss of dorsal and posterior fin structures, reduced eyes and 447 dorso-anterior head structures (Figure 1D). Not all structures are defective, for example 448 the cement gland appears unaffected (Figure 1D). These phenotypes are consistent 449 with the taf15 expression observed by ISH (Figure 1A, Stage 32). While morphological 450 defects were clear in stage 32 tadpoles, early embryos appeared fairly normal, although 451 gene expression changes were evident following TAF15 depletion (Figures 2-7). 452 Embryos injected with a translation-blocking MO containing five mismatches to the taf15 453 transcript (Mismatch) did not phenocopy the taf15 morphants, suggesting the effects of 454 the MO are specific to TAF15 depletion (Figure 1D). Following Z TAF15 depletion, 455 either by splice-blocking MO or CRISPR/Cas9-mutagenesis, embryos exhibit a milder 456 phenotype compared to the M+Z TAF15 depletion, with the most consistent phenotypes 457 between conditions being reduced eyes and dorso-anterior head structures (Figure 1D- 458 E). Importantly, embryos injected with Cas9 protein alone (Cas9 only) were similar to 459 uninjected embryos, again suggesting that taf15 sgRNA+Cas9 phenotype is specific to 460 the guide that mediates TAF15 depletion (Figure 1E). The discrepancy in the 461 phenotypes observed following M+Z or Z only TAF15 depletion suggests some 462 separable developmental roles for maternal and zygotic TAF15, as explored further 463 below. 464 465 To evaluate depletion efficiency, Western blot analysis was used to measure total Running Title: TAF15 regulates dorsoanterior neural development in Xenopus tropicalis Page 10

466 TAF15 following MO or sgRNA+Cas9 injection (Figure 1F). One possible reason for a 467 milder phenotype following Z TAF15 depletion (either MO or CRISPR/Cas9-mediated) is 468 likely the reduced TAF15 depletion efficiency in these conditions, as compared to M+Z 469 TAF15 depletion (Figure 1D,E,F). Indeed, none of the treatments eliminated TAF15 470 completely, illustrating the sensitivity of development to the dose of TAF15, confirmed 471 below by RNA sequencing (RNAseq). While we attempted to rescue the phenotypes 472 with injected taf15 mRNA, overexpression is also teratogenic, and we were unable to 473 find a dose that gave robust rescue. This is consistent with other experiments where 474 both increased and decreased expression of splicing regulators induces developmental 475 defects (Dichmann, Fletcher, & Harland, 2008; Iwasaki & Thomsen, 2014). Multiple 476 protein bands (most prominently at ~55kDa and ~60-65kDa) are observed for TAF15 by 477 Western blot (Figure 1A,E and Figure 4F). There are two known isoforms of taf15 in 478 Xenopus tropicalis. XM_012956707.2 comprises 1% of the taf15 transcript and 479 NM_001004806.1 comprises the remaining 99%; all of our depletion tools target the 480 latter 99%. TAF15 depletion affects all protein bands, suggesting they are specific to 481 TAF15 and targets of our depletion tools. As such, we suggest that multiple bands arise 482 from post-translational modifications. 483 484 Maternal TAF15 regulates splicing of developmental regulators. 485 486 In addition to the gross morphological defects that follow TAF15 depletion, changes in 487 gene expression were analyzed by single-embryo RNA sequencing (RNAseq). RNAseq 488 libraries were generated from stage 10 (gastrula stage) and 15 (neurula stage) embryos 489 injected with TAF15 translation-blocking MO (resulting in M+Z TAF15 depletion) (Figure 490 1C and 2A). The translation-blocking morphants were selected for sequencing as these 491 embryos displayed a more severe depletion phenotype (Figure 1D,E) and indicated a 492 consistently more robust TAF15 depletion (as assayed by Western blot Figure 1F). 493 TAF15 is a member of the FET family of proteins and the family member Fused in 494 liposarcoma (Fus) is necessary for the proper mRNA splicing of developmental 495 regulators in Xenopus (Dichmann & Harland, 2012). To understand how widespread, or 496 conserved, the roles of FET proteins are in splicing regulation in Xenopus we examined 497 changes in intron/exon usage and gene expression levels using the bioconductor 498 packages DEXseq and DESeq, respectively (Love, Huber, & Anders, 2014; Reyes, 499 Anders, & Huber, 2014) following M+Z TAF15 depletion (Figure 2A). 500 501 Following DEXseq, a two-fold threshold cutoff was applied to the differential exon usage 502 (DEU) gene candidates, followed by PANTHER GO-slim Biological Process analysis. In 503 stage 10 embryos, 228 transcripts were found to exhibit DEU, and 1,429 in stage 15 504 embryos (Supplemental Tables 1 & 2); 146 and 788 of which were assigned a 505 PANTHER GO-slim Biological Process respectively (Figure 2B). For the 100 genes with 506 affected differential exon usage at both stages 10 and 15, termed “stage persistent” 507 (Supplemental Table 3), we classified 86 genes with an identified PANTHER GO-slim 508 Biological Process (Figure 2B). Importantly, because DEXseq measures exon usage, it 509 is not an increase or decrease in gene expression that is measured in this “stage- 510 persistent” DEXseq cohort, but instead genes with conserved differences in exon usage 511 at both stages 10 and 15. Consistently we found enrichment for genes involved in 512 cellular and metabolic processes across all stages (Figure 2B). Using a database 513 designed to differentiate transcripts that are 1) “maternal”, meaning both maternally

Running Title: TAF15 regulates dorsoanterior neural development in Xenopus tropicalis Page 11

514 deposited and zygotically transcribed (present in the egg and expressed upon ZGA) 515 (e.g. fgfr4) from 2) “zygotic”, transcripts that are exclusively zygotically transcribed (e.g. 516 isl1), we were able to compare the percentage of observed target genes with DEUs that 517 have maternal expression to the percentage of all X. tropicalis genes with maternal 518 expression (database provided by the Rokhsar lab at U.C. Berkeley; not shown). The 519 “observed” category is comprised of the 100 stage-persistent genes that show 520 differential exon usage (Supplemental Table 3) and whether they have maternal and 521 zygotic (“maternal”) or exclusively zygotic (“zygotic”) transcript expression. Here we 522 observe 92% of these stage-persistent DEU target genes as transcripts known to have 523 both maternal and zygotic expression; with the remaining 8% composed of exclusively 524 zygotically expressed genes. The “expected” category represents the percentage of all 525 annotated X. tropicalis that have maternal and zygotic (“maternal”) or exclusively zygotic 526 (“zygotic”) transcript expression. Here we observe 65% of the X. tropicalis transcripts 527 have both maternal and zygotic expression with the remaining 35% exclusively 528 zygotically expressed (Figure 2C). We were therefore surprised to find 92% of the 529 stage-persistent genes to be transcripts that are both maternally deposited and 530 zygotically transcribed. These data suggest a preference for splice regulation of 531 transcripts that are present throughout development (from egg through ZGA) in this 532 M+Z TAF15 depletion condition, since only 65% of annotated X. tropicalis transcripts 533 are both maternally deposited and zygotically transcribed (Figure 2C). Alternatively, it is 534 also possible that intron retention within exclusively zygotic transcripts is less abundant 535 than expected in the M+Z TAF15 depleted embryos. Interestingly, of the 100 stage- 536 persistent genes, 83 exhibit one intron retention, 13 have two introns retained, and four 537 have more than two introns retained (Figure 2C). Of the 100 “stage-persistent” DEU 538 genes, we found 53 distributed throughout the top 100 DEU genes of the stage 10 539 embryo (as sorted by adjusted P value) suggesting that this approach of looking for 540 conserved DEUs between stages 10 and 15 is robust for finding splicing events with an 541 early and lasting effect throughout development. To ensure we further investigated 542 genes with true DEUs, we used qRT-PCR to validate a subset of the DEU candidates 543 that were found by RNAseq of TAF15-depleted embryos to have robustly retained 544 introns (of which there were 29/100; Supplemental Table 3, column “Retained introns as 545 visualized by DEXseq gene modeling”). 9/12 (75%) of these DEXseq RNA-seq results 546 were validated by qRT-PCR (Figure 2D). These 12 cases of intron retention correspond 547 to a total of 11 genes due to the validation of two introns belonging to dgka. 548 Surprisingly, this qRT-PCR data lead to the finding that intron retention is exclusively 549 found in embryos following M+Z TAF15 depletion; intron retention was never observed 550 by qRT-PCR in embryos depleted of zygotic TAF15 or injected with mismatch MO 551 (Figure 2D). Additionally, we show that the splice-blocking morpholino is specific to Z 552 TAF15 depletion only and does not have an effect on maternally deposited TAF15 553 (Supplemental Figure 1); further supporting our hypothesis that splice changes are due 554 to dose dependence of TAF15, which requires translation of maternal taf15 mRNA. 555 556 Depletion of M+Z TAF15 leads to intron retention in fgfr4 557 558 A two-fold expression cutoff was applied to the DEXseq and DEseq results (except for 559 stage 10 DESeq) and the candidate genes were sorted by their adjusted P (padj) 560 values. DEXseq results indicate fgfr4 intron 1 to be the 1st and 7th hit in stage 10 and 15 561 embryos respectively (Supplemental tables 1 and 2). Numerous top-ranking candidate

Running Title: TAF15 regulates dorsoanterior neural development in Xenopus tropicalis Page 12

562 DEUs were visualized using the Integrative Genomics Viewer (IGV) and DEXseq DEU 563 models at both stage 10 and stage 15 to gauge the significance of each DEU. Of these 564 top candidates, fgfr4 displayed the most robust intron retention (Supplemental figures 2 565 and 3). Additionally, FGFR4 was an intriguing candidate because it is known that 566 FGFRs are alternatively spliced. Examining the levels of fgfr4 expression, DESeq 567 results indicate the fgfr4 transcript to be the 20th and 133rd hit in stage 15 and 10 568 embryos respectively (Supplemental tables 4 and 5). These results were validated by 569 qRT-PCR in M+Z TAF15-depleted single embryos using fgfr4 intron 1-specific primers; 570 fgfr4 intron 1 was assayed across all depletion conditions yet was not detected outside 571 of the M+Z TAF15 depletion condition (Figure 3B). To confirm that fgfr4 intron 1 qRT- 572 PCR results were specific to retention (as suggested by visualization by IGV; Figure 3A) 573 and not simply increased due to an overall upregulation of fgfr4 pre-mRNA, the fgfr4 574 intron 1 expression levels (fgfr4 intron 1-specific primers, red primer set Figure 3A) were 575 normalized to the total fgfr4 expression levels (Figure 3B; fgfr4 total transcript-specific 576 primers, blue primer set Figure 3A). As a result of normalizing fgfr4 intron 1 expression 577 to total fgfr4 expression, conditions that do not retain fgfr4 intron 1 will have a relative 578 expression equal to 1 within this analysis (Figure 3B). Consistent with our qRT-PCR 579 findings in Figure 2D, intron retention is not observed following zygotic TAF15 depletion 580 by morpholino or sgRNA/Cas9 (Figure 3B). Interestingly, however, loss of zygotic 581 TAF15 (both by MO and CRISPR/Cas9) does lead to an overall reduction in total fgfr4 582 expression (Figure 3C); these results contrast sharply with increased expression and 583 intron retention of fgfr4 expression following M+Z taf15 depletion (Figure 3C). 584 Importantly, fgfr4 expression is unaffected following injection of control morpholino or 585 Cas9 protein alone, suggesting that the changes in fgfr4 expression are specific to 586 TAF15 depletion (Figure 3B-C). 587 588 We next used RNA in situ hybridization (ISH) to total fgfr4 to determine how the 589 changes observed by qRT-PCR and RNAseq following TAF15 depletion may affect the 590 expression pattern of fgfr4. M+Z depletion of TAF15 leads to an increase and expansion 591 of the lateral domain of fgfr4 expression just outside the neural plate (Figure 4A). We 592 hypothesize that the intron-containing fgfr4 transcripts fail to undergo degradation 593 (necessitating normalization of intron levels to total transcript to avoid artificially inflating 594 the amount of retained intron measured by qRT-PCR in Figure 3B) and are largely 595 responsible for this diffuse increase in expression and that these transcripts are non- 596 functional forms of fgfr4; indeed, M+Z TAF15 depletion results in reduced FGFR4 597 protein levels as measured by western blot (Figure 4F). Following zygotic depletion of 598 TAF15 by MO or sgRNA/Cas9 we observe a decrease in fgfr4 transcript and protein 599 (assayed for morphants only) expression (Figure 4B, D, F). Importantly, fgfr4 expression 600 is unaffected following injection of control morpholino or Cas9 protein alone, suggesting 601 that these results are specific to TAF15 depletion (Figure 4C, E). These results are 602 consistent with the qRT-PCR data and suggest that the fgfr4 transcripts are universally 603 affected in each TAF15-depleted embryo, as the pattern of expression is largely 604 unaffected (lateral expansion of fgfr4 adjacent to the neural plate is observed in Z 605 TAF15 depleted morphant embryos suggesting a modest delay in convergence as we 606 observe complete neural tube closure). These findings further show that both M+Z and 607 Z TAF15 MO depletion lead to a loss of FGFR4 protein (Figure 4F) and we hypothesize 608 that this reduction is achieved through two different mechanisms: M+Z TAF15 depletion 609 leads to intron retention in fgfr4, resulting in an early stop codon (data not shown), and

Running Title: TAF15 regulates dorsoanterior neural development in Xenopus tropicalis Page 13

610 in reduced FGFR4, whereas Z TAF15 depletion leads to the reduction of total fgfr4 611 mRNA expression, also resulting in reduced FGFR4. 612 613 Lastly, after determining that total FGFR4 expression is reduced following both M+Z and 614 Z-only TAF15 MO depletion, we next determined if downstream targets of FGFR4 were 615 affected. Indeed, the two midbrain/hindbrain markers, pax2 and engrailed 2, show 616 reduction both by ISH or qRT-PCR across all TAF15 depletion conditions (Figure 4G-L) 617 (Hongo, Kengaku, & Okamoto, 1999). These results indicate that M+Z and Z-only 618 TAF15 depletion lead to the downregulation of FGFR4, but through independent 619 mechanisms of transcriptional regulation. 620 621 We previously showed that 83% of “stage-persistent” DEUs are characterized by the 622 retention of a single intron (Figure 2C); this is consistent with what is observed in fgfr4. 623 A broader analysis determined if first introns are preferred targets or if fgfr4 is an 624 exception. In examining the top 10 genes with affected DEUs (as sorted by adjusted P 625 value) at both stages 10 and 15, we found the single retained introns are not restricted 626 to first introns, but are dispersed in different transcripts (Supplemental Figures 2 and 3; 627 Supplemental Table 7). Therefore, retention of a single intron is the most common 628 characteristic of DEU following M+Z TAF15 depletion (13/20 cases) within this cohort, 629 contrasting with the effects of Fus depletion, which affects all introns of DEU genes. 630 631 Measuring changes in gene expression following TAF15 depletion. 632 633 It is unknown if and how the FET family of proteins regulates levels of gene expression 634 in Xenopus. Our data thus far suggests that TAF15 controls both mRNA splicing and 635 overall levels of a subset of RNAs in the developing embryo. Concurrent with analyzing 636 how TAF15 depletion affects changes in intron/exon usage (DEXseq), we also 637 examined levels of transcript abundance using DESeq (Figure 5). 638 639 Following DESeq, in an effort to focus on a few candidate genes that exhibit differential 640 expression following TAF15 depletion, a two-fold threshold cutoff was applied to the 641 differentially expressed gene candidates, followed by PANTHER GO-slim Biological 642 Process analysis. In stage 15 embryos, 2,094 transcripts were found to exhibit a two- 643 fold increase or decrease in gene expression (Supplemental Table 6), 1,235 of which 644 were assigned a PANTHER GO-slim Biological Process (Figure 5A; 773 total genes 645 with decreased expression plus 462 total genes with increased expression). Our taf15 646 ISH data suggest that taf15 has a specific expression pattern, therefore, we looked 647 more closely at those genes that fall under the PANTHER GO-slim Developmental 648 process (Figure 5A, asterisk); 207 genes were found to be differentially expressed 649 within this classification (Figure 5B; combined Developmental Process results from 650 Figure 5A). More specifically, our taf15 ISH data suggests that taf15 is expressed in the 651 ectoderm at stage 15 (Figure 1A, Stage 15), therefore the subset of transcripts 652 annotated with the Ectoderm development classification were more closely examined 653 (Figure 5B, C). One family of genes in this category, that increases in expression 654 following M+Z TAF15 depletion, is the VENT family of homeodomain transcription 655 factors (ventx), suggesting a role for TAF15 suppression of ventx expression in the 656 ectoderm (Figure 5C). Interestingly, ventx genes act in a positive feedback loop with the 657 bone morphogenetic proteins (BMP), which specify the ventral domain of the Xenopus

Running Title: TAF15 regulates dorsoanterior neural development in Xenopus tropicalis Page 14

658 embryo (Onichtchouk, Glinka, & Niehrs, 1998; Sander et al., 2007). Because taf15 is 659 expressed in the future dorsal domain of the gastrula (Figure 1A), we were intrigued to 660 find a gene family (vent) that functions in ventral tissue development, and is upregulated 661 upon TAF15 depletion. This suggests an early and lasting regulatory relationship 662 between taf15 and ventx. Of the four ventx paralogs upregulated following M+Z TAF15 663 depletion, ventx2.1 acts upstream of ventx1.1, ventx2.2, and ventx3.1 therefore we 664 investigate the relationship between ventx2.1 and taf15 further (Schuler-Metz, Knöchel, 665 Kaufmann, & Knöchel, 2000). 666 667 TAF15 regulates dorsoanterior development by repressing ventx2.1. 668 669 To better understand the relationship between taf15 and the ventx genes, we analyzed 670 their expression patterns further. In situ hybridization revealed a subset of 671 complementary expression patterns between taf15 and ventx2.1, consistent with 672 inhibition of ventx2.1 by taf15 (Figure 6A-H). At stage 10, taf15 is expressed in the 673 embryonic domain that will give rise to future dorsal structures whereas ventx2.1 674 expression is markedly absent from this region and is instead expressed in the 675 embryonic domain that will give rise to future ventral structures (Figure 6A-D). ventx2.1 676 and taf15 continue their complementary expression into the neurula stage (Figure 6E- 677 H). To visualize the extent of complementary expression, medial cross sections along 678 the dorsoventral axis were made through stage 15 embryos following ISH. These 679 sections reveal that taf15 is strongly expressed throughout the neuroectoderm and in 680 both the epithelial and sensorial layers of the ectoderm whereas venx2.1 is strongly 681 expressed in the underlying lateral plate mesoderm and only faintly expressed in a 682 portion of the neuroectoderm (Figure 6E’, F’). One domain where taf15 and ventx2.1 683 exhibit overlapping expression is in the forebrain (Figure 6E-H). 684 685 Having established by RNAseq that ventx2.1 expression increases with M+Z TAF15 686 depletion, and that taf15 and ventx2.1 have some complementary expression patterns, 687 we examined ventx2.1 expression in situ in TAF15-depleted embryos. Embryos with 688 depleted M+Z TAF15 showed increased ventx2.1 expression in regions that normally 689 have strong taf15 and weak ventx2.1 expression, specifically below the lateral 690 neuroectoderm and prospective epidermal region (Figure 6 E,F,J,K (white bracket) and 691 L). Additionally, ventx2.1 expression increases in the region of the forebrain where taf15 692 and ventx2.1 are coexpressed (Figure 6 G,H,I (arrowhead) and L). Taken together, 693 these data suggest that TAF15 suppresses ventx2.1; and presumably does so 694 indirectly, since the genes are expressed in adjacent germ layers. The increased level 695 of ventx2.1 is consistent with the gross phenotype of reduced head structures observed 696 in older TAF15-depleted embryos (Figure 1D,E & Figure 6L, red outline). 697 698 TAF15 regulates ventx2.1 throughout early embryogenesis. 699 700 RNAseq data show that M+Z TAF15 depletion results in increased ventx2.1 expression 701 by neurula stage 15 but not at the earlier gastrula stage 10 (<2 fold change) (Figure 7A). 702 However, qRT-PCR results, show ventx2.1 expression significantly increased at both 703 stage 10 and 15 following M+Z TAF15 depletion (Figure 7B); consistent with the 704 complementary taf15 and ventx2.1 expression patterns observed by ISH at stage 10 705 (Figure 6A-D). Following Z-only TAF15 depletion, we observe that ventx2.1 expression

Running Title: TAF15 regulates dorsoanterior neural development in Xenopus tropicalis Page 15

706 is unaffected at stage 10 by qRT-PCR; however, by stage 15 the Z-only TAF15 707 depleted embryos phenocopy those that are M+Z TAF15 depleted (Figure 7B). We 708 hypothesize that the effects of Z-only TAF15 depletion on embryonic development are 709 not observed as early as stage 10 because zygotic genome transcription has only just 710 begun (possibly not giving the zygotic genome-targeting MO enough time to act), 711 furthermore, maternal TAF15 could still be present and able to suppress the 712 transcription of ventx2.1. Embryos injected with a mismatch MO to the translation start 713 site of taf15 do not exhibit changes in ventx2.1 expression at either stage, suggesting 714 that the changes in gene expression observed in M+Z and Z-only TAF15 depleted 715 embryos are specific to morpholino-mediated TAF15 depletion (Figure 7B, E, J). 716 717 To ensure that the increase in ventx2.1 expression is not a morpholino-specific effect, 718 we measured ventx2.1 expression following TAF15 depletion with the CRISPR/Cas9 719 system. qRT-PCR data for ventx2.1 expression in stage 15 sgRNA + Cas9-injected 720 embryos phenocopies that observed in M+Z and Z-only TAF15 depleted embryos; 721 importantly, injection of Cas9 alone does not affect ventx2.1 expression level (Figure 722 7B). In addition to quantifying the changes in ventx2.1 expression by qRT-PCR, the 723 embryonic expression pattern of ventx2.1 was also compared between MO and 724 CRISPR/Cas9 conditions. We find that the expression pattern of ventx2.1 following 725 sgRNA taf15 + Cas9 phenocopies that of M+Z and Z-only TAF15 depleted embryos 726 (Figure 7C-L). Taken together, these qRT-PCR and ISH data suggest that changes in 727 ventx2.1 expression is specific to TAF15 depletion and that both maternal and zygotic 728 TAF15 play a role in suppressing the expression of ventx2.1 throughout development. 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753

Running Title: TAF15 regulates dorsoanterior neural development in Xenopus tropicalis Page 16

754 755 DISCUSSION 756 757 Previous to our investigation, few studies have focused on the role of TATA-binding 758 protein associated factor 15 (TAF15) in development. TAF15 is not considered a 759 canonical TATA-binding protein associated factor (TAF) as it is not associated with all 760 human TFIID complexes and has no ortholog in invertebrate species (Ballarino et al., 761 2012). However, it is this non-ubiquitous association with the core transcriptional 762 machinery that interests us as this supports the hypothesis that TAF15 may be selective 763 in the transcripts it regulates and, as a result, may have specific roles in development. 764 TAF15 would not be the first TAF to be shown to have a specific role in development, 765 for example, TAF3 is required for endoderm lineage differentiation and preventing the 766 premature specification of neurectoderm and mesoderm in embryonic stem cells (Liu et 767 al., 2011). 768 The analysis presented here demonstrates that TAF15 is deposited maternally in 769 Xenopus eggs and is later expressed zygotically. We further show that TAF15 has an 770 enriched expression pattern within the developing embryo with maternal taf15 771 preferentially in the animal hemisphere and zygotic taf15 enriched dorsally during 772 gastrulation, throughout the neural ectoderm during neurulation, and in dorsoanterior 773 tissues during tailbud and tadpole stages. Consistent with the expression pattern of 774 taf15 we find that embryos depleted of taf15 have defects in head structures such as 775 reduced fore/midbrain and eyes by the early tadpole stage. Interestingly, by employing 776 RNA sequencing (RNAseq) to understand the role of TAF15 in development we made 777 the surprising discovery that maternal and zygotic TAF15 exhibit different mechanisms 778 by which they regulate FGFR4 expression: the depletion of maternal and zygotic TAF15 779 together downregulates FGFR4 expression at the post-transcriptional level via the 780 retention of a single fgfr4 intron, while depletion of zygotic TAF15 alone downregulates 781 FGFR4 expression through the reduction of total fgfr4 transcript. Additionally, we find 782 that TAF15 is required to repress ventx2.1 from dorsal and neural ectodermal tissues 783 and that taf15 and ventx2.1 exhibit a complementary expression pattern through 784 gastrulation and neurulation. The data presented here suggest TAF15 plays an integral 785 and pleiotropic role in development of the dorsoanterior neural tissues and further 786 suggest that the mechanism of gene regulation by TAF15 is target-dependent and 787 subject to the milieu of factors that are present at different times of development, likely 788 due to the presence of specific co-factors required for activity (Figure 8). 789 790 It is not unexpected that TAF15 plays a pleiotropic role in Xenopus development as FET 791 proteins are associated with regulating numerous cellular activities including: cell 792 proliferation, cell cycling, cell death, transcription, splicing, microRNA processing, RNA- 793 transport, signaling, and maintenance of genomic integrity (Andersson et al., 2008; 794 Ballarino et al., 2012; Gregory et al., 2004; Shiohama, Sasaki, Noda, Minoshima, & 795 Shimizu, 2007). Furthermore, we expected that splicing could be a shared mechanism 796 of gene regulation between Fus and TAF15 in Xenopus development as studies using 797 photoactivatable ribonucleoside-enhanced cross-linking and immunoprecipitation (PAR- 798 CLIP) found that FET proteins predominantly bind to intronic regions as well as the 799 3’UTR of genes (Hoell et al., 2011). However, we are surprised by how differently the 800 two FET family members, Fus and TAF15, affect Xenopus development; embryos Running Title: TAF15 regulates dorsoanterior neural development in Xenopus tropicalis Page 17

801 depleted of Fus fail to undergo gastrulation, due to the retention of all introns in a subset 802 of target genes required for mesoderm differentiation and epithelial adhesion (Dichmann 803 & Harland, 2012), whereas embryos depleted of TAF15 survive into the tadpole stage 804 and exhibit defects specific to dorsoanterior head development due in part to 805 dysregulation of fgfr4 and ventx2.1, and in contrast to Fus, TAF15 depletion affects a 806 subset of specific introns. 807 808 TAF15 provides an example of a gene product with two different mechanisms (post- 809 transcriptional or transcriptional) by which to regulate the expression of the same target 810 gene (FGFR4). Furthermore, we have shown that the mode of regulation is dependent 811 on whether TAF15 is maternally deposited or zygotically transcribed. While there are 812 classic examples of genes having separable maternal and zygotic developmental roles 813 (e.g. β-catenin) (Heasman et al., 1994; Heasman et al., 2000), we are unaware of a 814 case where the mechanism of regulatory action changes while the target remains the 815 same. 816 817 Classical RNA splicing can be separated in two functional groups, constitutive and 818 alternative splicing. Constitutive splicing is the process by which introns are removed 819 (spliced), stitching together exons in the same order that they are found in the genome, 820 producing one gene product (Boutz, Bhutkar, & Sharp, 2015; Pandya-Jones, 2011; 821 Perales & Bentley, 2009). Alternative splicing refers to the process by which exons of a 822 gene may be included or excluded, producing numerous gene products (isoforms) and 823 increasing gene product diversity and complexity (Black, 2003; Grabowski & Black, 824 2001). Both constitutive and alternative splicing occur co-transcriptionally, prior to the 825 transcriptional termination and polyadenylation of pre-mRNAs (Pandya-Jones & Black, 826 2009). In addition to co-transcriptional splicing, there is also post-transcriptional splicing. 827 It has been shown that the retention of individual introns in poly-adenylated pre-mRNAs 828 serves as a mechanism for controlling gene expression; transcripts with retained introns 829 will remain in the nucleus, preventing translation of the transcript, but following a cellular 830 signal (e.g. osmotic or heat stress), the intron is excised and the protein is quickly 831 translated (Boutz et al., 2015; Ninomiya, Kataoka, & Hagiwara, 2011). Here we closely 832 studied fgfr4, a transcript that is both maternally deposited and zygotic transcribed and 833 observe post-transcriptional splicing defects (in fgfr4 as well as other targets) following 834 the depletion of maternal and zygotic TAF15. Our RNAseq data, generated from post- 835 ZGA embryos (stages 10 and 15) support a model whereby translation of maternally 836 deposited taf15 is required for the proper post-transcriptional splicing of nascent zygotic 837 pre-mRNA transcripts during the time of zygotic genome activation. Indeed, in looking at 838 early RNA seq datasets (Xenbase) we find no evidence that any of the maternal 839 transcript (e.g. fgfr4) is incompletely spliced. 840 841 In the embryonic environment where zygotic genome activation occurs, and therefore 842 active transcription is taking place, we propose that both maternally deposited TAF15, 843 and TAF15 translated post-fertilization aids in splicing the new (zygotic) transcripts (e.g. 844 fgfr4, isl1, and others; Supplemental Figures 2 and 4), resulting in unspliced transcripts 845 at stage 10 after TAF15 depletion, but that zygotic TAF15 acts more classically and 846 likely associates with the core promoter, using either its N-terminal low complexity 847 domain or RNA-binding domains to bind the C-terminal domain of RNA pol II, and 848 regulates transcription of specific targets (Figure 8). Upon zygotic genome activation

Running Title: TAF15 regulates dorsoanterior neural development in Xenopus tropicalis Page 18

849 (production of nascent transcripts), zygotic TAF15 may associate with the core promoter 850 to regulate the expression of zygotically transcribed targets (e.g. fgfr4) (Figure 8). Our 851 data also show that there is not always a discrepancy in maternal-zygotic TAF15 target 852 regulation. In the case of ventx2.1, which is expressed zygotically (and not maternally 853 deposited), we do not observe a splicing defect following TAF15 depletion, and find that 854 both maternal and zygotic TAF15 depletion results in increased ventx2.1 expression. 855 856 We now know that at least one member of the FET family of atypical RNA-binding 857 proteins, TAF15, is required to regulate dorsoventral patterning 858 in Xenopus. Xenopus embryos depleted of TAF15 have a phenotype similar to that 859 which we observe when embryos are depleted of the BMP-antagonist chordin, or pairs 860 of BMP antagonists (Reversade et al., 2005; Khokha et al., 2005). 861 Both chordin and taf15 depletion results in ventralized embryos with reduced head and 862 eye structures as well as reduced dorsal and posterior fin and tail structures. Just as 863 with TAF15 depletion, chordin-depleted embryos have a relatively normal cement gland 864 and increased ventral tissue. Interestingly, according to our RNAseq data, TAF15- 865 depleted embryos do not decrease chordin (chrd) expression, in fact we observe a two- 866 fold increase at stage 15 (Supplemental Table 4 and 6). However, rather than regulate 867 BMPs directly, we suggest a model where TAF15 is needed to repress vent genes and 868 thereby disrupts the BMP/Vent positive feedback loop (Figure 6M) (Schuler-Metz et al., 869 2000).These data clearly illustrate the role of TAF15 in regulating dorsoventral 870 patterning. We propose a model where TAF15 represses ventx2.1 from the dorsal 871 marginal zone of the gastrula and this repression continues through neurulation; we see 872 TAF15 repressing ventx2.1 from dorsal neural ectodermal tissue. Without knowing if the 873 ventx genes are direct transcriptional targets of TAF15, we cannot conclude if the 874 function of TAF15 is to repress ventx expression or to activate a repressor of ventx 875 expression. However, the RNAseq results show that there is no intron retention in the 876 ventx genes which supports the conclusion that the RNA splicing activity of TAF15 does 877 not control the TAF15-dependent repression of ventx2.1. 878 879 Interestingly, the human Vent-like homeobox gene VENTX, a putative homolog of the 880 Xenopus ventx2 gene is aberrantly expressed in CD34+ cells of acute myeloid leukemia 881 patients (Rawat et al., 2010). Furthermore, the leukemia-associated TAF15 fusion 882 protein, TAF15-CIZ/NMP4, is found in acute myeloid leukemia (Alves et al., 2009). 883 Although ventx has been lost in mouse, the function of ventx in repressing dorsal fates 884 is well conserved between fish and frogs (Imai et al., 2001; Rawat et al., 2010). Given 885 the relationship we have observed of increased ventx expression following TAF15 886 depletion, the coincidence of TAF15 dysfunction and increased VENTX in acute myeloid 887 leukemia, and the fact that Ventx is required for proper mesenchyme and blood 888 differentiation in Xenopus, it is possible that TAF15-dependent negative regulation of 889 ventx is a conserved mechanism (Onichtchouk et al., 1998). 890 891 In summary, the data presented here show that a target gene (fgfr4) is regulated 892 through two different molecular mechanisms (post-transcriptionally or transcriptionally) 893 depending on the mechanism of TAF15 depletion (maternal + zygotic or zygotic-only). 894 We have also demonstrated that while the effects of TAF15 on development are 895 pleiotropic, consistent with our observations that embryos exhibit reduced head 896 structures following TAF15 depletion, we have identified two specific pathways by which

Running Title: TAF15 regulates dorsoanterior neural development in Xenopus tropicalis Page 19

897 TAF15 regulates dorsoanterior brain development (fgfr4 and ventx2.1). Furthermore, we 898 have demonstrated that the FET family of atypical RNA-binding proteins do not act 899 redundantly in regulating Xenopus development; TAF15 and Fus-depleted embryos 900 exhibit very distinct phenotypes as well as mechanisms of gene expression regulation. 901 Our findings of non-overlapping transcriptional effects by TAF15 and Fus in Xenopus 902 supports the idea that the disease consequences resulting from chromosomal 903 translocations involving the FET family of proteins are unlikely to be common 904 transcriptional effects, but are more likely to result from the similar cellular aggregates of 905 mutant proteins (Scekic‐Zahirovic et al., 2016; Sharma et al., 2016; Svetoni et al., 906 2016). 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944

Running Title: TAF15 regulates dorsoanterior neural development in Xenopus tropicalis Page 20

945 946 Author Contributions 947 C.S.D. and RMH) designed and performed the experiments and wrote and edited the 948 manuscript. D.D.S. assisted in bioinformatic analysis and laid the groundwork for 949 studying FET proteins in Xenopus. C.R.T.E. and Y.X. performed embryo injections and 950 imaging for revision experiments. R.M.H. conceived of the project and edited the 951 manuscript. 952 953 Acknowledgements 954 We would like to thank Marta Truchado Garcia and Edivinia Pangilinan of the Harland 955 lab at the University of California, Berkeley for organizing reagents for revision 956 experiments. We would also like to thank Helen Willsey of Matthew State’s lab for 957 providing Xenopus tropicalis embryos for revision experiments. We would like to thank 958 Jim Boonyaratanakornkit and Matthew Gray of Justin Taylor’s lab at the Fred 959 Hutchinson Cancer Research Center for their time and Western blot reagents used for 960 revision experiments. 961 962 Competing Interested 963 The authors declare no competing financial interests. 964 965 966 REFERENCES 967 968 Alves, J., Wurdak, H., Garay-Malpartida, H. M., Harris, J. L., Occhiucci, J. M., Belizário, 969 J. E., & Li, J. (2009). TAF15 and the leukemia-associated fusion protein TAF15- 970 CIZ/NMP4 are cleaved by caspases-3 and -7. Biochemical and Biophysical 971 Research Communications, 384(4), 495–500. 972 https://doi.org/10.1016/j.bbrc.2009.05.009 973 Andersson, M. K., Ståhlberg, A., Arvidsson, Y., Olofsson, A., Semb, H., Stenman, G., … 974 Åman, P. (2008). The multifunctional FUS, EWS and TAF15 proto-oncoproteins 975 show cell type-specific expression patterns and involvement in cell spreading and 976 stress response. BMC Cell Biology, 9(1), 37. https://doi.org/10.1186/1471-2121-9- 977 37 978 Ballarino, M., Jobert, L., Dembélé, D., de la Grange, P., Auboeuf, D., & Tora, L. (2012). 979 TAF15 is important for cellular proliferation and regulates the expression of a 980 subset of cell cycle genes through miRNAs. Oncogene, (May), 1–10. 981 https://doi.org/10.1038/onc.2012.490 982 Black, D. L. (2003). Mechanisms of Alternative Pre-Messenger RNA Splicing. Annual 983 Review of Biochemistry, 72(1), 291–336. 984 https://doi.org/10.1146/annurev.biochem.72.121801.161720 985 Boutz, P. L., Bhutkar, A., & Sharp, P. A. (2015). Detained introns are a novel, 986 widespread class of post-transcriptionally spliced introns. Genes & Development, 987 29(1), 63–80. https://doi.org/10.1101/gad.247361.114 988 Crozat, A., Åman, P., Mandahl, N., & Ron, D. (1993). Fusion of CHOP to a novel RNA- 989 binding protein in human myxoid liposarcoma. Nature, 363(6430), 640–644. 990 https://doi.org/10.1038/363640a0 991 Delattre, O., Zucman, J., Plougastel, B., Desmaze, C., Melot, T., Peter, M., … Thomas, 992 G. (1992). Gene fusion with an ETS DNA-binding domain caused by chromosome

Running Title: TAF15 regulates dorsoanterior neural development in Xenopus tropicalis Page 21

993 translocation in human tumours. Nature, 359(6391), 162–165. 994 https://doi.org/10.1038/359162a0 995 Dichmann, D. S., Fletcher, R. B., & Harland, R. M. (2008). Expression cloning in 996 Xenopus identifies RNA-binding proteins as regulators of embryogenesis and 997 Rbmx as necessary for neural and muscle development. Developmental 998 Dynamics: An Official Publication of the American Association of Anatomists, 999 237(7), 1755–1766. https://doi.org/10.1002/dvdy.21590 1000 Dichmann, D. S., & Harland, R. M. (2012). fus/TLS orchestrates splicing of 1001 developmental regulators during gastrulation. Genes & Development, 26(12), 1002 1351–1363. https://doi.org/10.1101/gad.187278.112 1003 Grabowski, P. J., & Black, D. L. (2001). Alternative RNA splicing in the nervous system. 1004 Progress in Neurobiology, 65(3), 289–308. https://doi.org/10.1016/S0301- 1005 0082(01)00007-7 1006 Gregory, R. I., Yan, K.-P., Amuthan, G., Chendrimada, T., Doratotaj, B., Cooch, N., & 1007 Shiekhattar, R. (2004). The Microprocessor complex mediates the genesis of 1008 microRNAs. Nature, 432(7014), 235–240. https://doi.org/10.1038/nature03120 1009 Heasman J, Crawford A, Goldstone K, Garner-Hamrick P, Gumbiner B, McCrea P, 1010 Kintner C, Noro CY, W. C. (1994). Overexpression of cadherins and 1011 underexpression of?-catenin inhibit dorsal mesoderm induction in early Xenopus 1012 embryos. Cell, 79(5), 791–803. https://doi.org/10.1016/0092-8674(94)90069-8 1013 Heasman, J., Kofron, M., & Wylie, C. (2000). βCatenin Signaling Activity Dissected in 1014 the Early Xenopus Embryo: A Novel Antisense Approach. Developmental Biology, 1015 222(1), 124–134. https://doi.org/10.1006/dbio.2000.9720 1016 Hicks, G. G., Singh, N., Nashabi, A., Mai, S., Bozek, G., Klewes, L., … Ruley, H. E. 1017 (2000). Fus deficiency in mice results in defective B-lymphocyte development and 1018 activation, high levels of chromosomal instability and perinatal death. Nature 1019 Genetics, 24(2), 175–179. https://doi.org/10.1038/72842 1020 Hoell, J. I., Larsson, E., Runge, S., Nusbaum, J. D., Duggimpudi, S., Farazi, T. a, … 1021 Tuschl, T. (2011). RNA targets of wild-type and mutant FET family proteins. Nature 1022 Structural & Molecular Biology, 18(12), 1428–1431. 1023 https://doi.org/10.1038/nsmb.2163 1024 Hongo, I., Kengaku, M., & Okamoto, H. (1999). FGF signaling and the anterior neural 1025 induction in Xenopus. Developmental Biology, 216(2), 561–581. 1026 https://doi.org/10.1006/dbio.1999.9515 1027 Imai, Y., Gates, M. a, Melby, a E., Kimelman, D., Schier, a F., & Talbot, W. S. (2001). 1028 The homeobox genes vox and vent are redundant repressors of dorsal fates in 1029 zebrafish. Development (Cambridge, England), 128(12), 2407–2420. 1030 Iwasaki, Y., & Thomsen, G. H. (2014). The splicing factor PQBP1 regulates 1031 mesodermal and neural development through FGF signaling. Development 1032 (Cambridge), 141(19), 3740–3751. https://doi.org/10.1242/dev.106658 1033 Kapeli, K., Pratt, G. A., Vu, A. Q., Hutt, K. R., Martinez, F. J., Sundararaman, B., … 1034 Yeo, G. W. (2016). Distinct and shared functions of ALS-associated proteins TDP- 1035 43, FUS and TAF15 revealed by multisystem analyses. Nature Communications, 7. 1036 https://doi.org/10.1038/ncomms12143 1037 Kato, M., Han, T. W., Xie, S., Shi, K., Du, X., Wu, L. C., … McKnight, S. L. (2012). Cell- 1038 free formation of RNA granules: low complexity sequence domains form dynamic 1039 fibers within hydrogels. Cell, 149(4), 753–767. 1040 https://doi.org/10.1016/j.cell.2012.04.017

Running Title: TAF15 regulates dorsoanterior neural development in Xenopus tropicalis Page 22

1041 Khokha, M. K., Chung, C., Bustamante, E. L., Gaw, L. W. K., Trott, K. a., Yeh, J., … 1042 Grammer, T. C. (2002). Techniques and probes for the study of Xenopus tropicalis 1043 development. Developmental Dynamics, 225(4), 499–510. 1044 https://doi.org/10.1002/dvdy.10184 1045 Khokha, M. K., Yeh, J., Grammer, T. C., & Harland, R. M. (2005). Depletion of three 1046 BMP antagonists from Spemann’s organizer leads to a catastrophic loss of dorsal 1047 structures. Developmental Cell, 8(3), 401–411. 1048 https://doi.org/10.1016/j.devcel.2005.01.013 1049 King, O. D., Gitler, A. D., & Shorter, J. (2012). The tip of the iceberg: RNA-binding 1050 proteins with prion-like domains in neurodegenerative disease. Brain Research, 1051 1462, 61–80. https://doi.org/10.1016/j.brainres.2012.01.016 1052 Kovar, H. (2011). Dr. Jekyll and Mr. Hyde: The Two Faces of the FUS/EWS/TAF15 1053 Protein Family. Sarcoma, 2011(Table 1), 837474. 1054 https://doi.org/10.1155/2011/837474 1055 Li, H., Watford, W., Li, C., Parmelee, A., Bryant, M. a., Deng, C., … Lee, S. B. (2007). 1056 Ewing sarcoma gene EWS is essential for meiosis and B lymphocyte development. 1057 Journal of Clinical Investigation, 117(5), 1314–1323. 1058 https://doi.org/10.1172/JCI31222 1059 Liu, Z., Scannell, D. R., Eisen, M. B., & Tjian, R. (2011). Control of embryonic stem cell 1060 lineage commitment by core promoter factor, TAF3. Cell, 146(5), 720–731. 1061 https://doi.org/10.1016/j.cell.2011.08.005 1062 Love, M. I., Huber, W., & Anders, S. (2014). Moderated estimation of fold change and 1063 dispersion for RNA-seq data with DESeq2. https://doi.org/10.1101/002832 1064 Martini, A., Starza, R. La, Janssen, H., Corveleyn, A., Somers, R., Aventin, A., … 1065 Marynen, P. (2002). Advances in Brief Recurrent Rearrangement of the Ewing ’ s 1066 Sarcoma Gene , EWSR1 , or Its Homologue , TAF15 , with the Transcription Factor 1067 CIZ / NMP4 in Acute Leukemia. Cancer, 1, 5408–5412. 1068 Neumann, M., Bentmann, E., Dormann, D., Jawaid, A., DeJesus-Hernandez, M., 1069 Ansorge, O., … Mackenzie, I. R. A. (2011). FET proteins TAF15 and EWS are 1070 selective markers that distinguish FTLD with FUS pathology from amyotrophic 1071 lateral sclerosis with FUS mutations. Brain, 134(9), 2595–2609. 1072 https://doi.org/10.1093/brain/awr201 1073 Nieuwkoop, P.D. and Faber, J. (1994). Normal Table of Xenopus Laevis (Daudin) : a 1074 system- atical and chronological survey of the development from the fertilized egg 1075 till the end of metamorphosis. New York: Garland Publishing. 1076 Ninomiya, K., Kataoka, N., & Hagiwara, M. (2011). Stress-responsive maturation of 1077 Clk1/4 pre-mRNAs promotes phosphorylation of SR splicing factor. The Journal of 1078 Cell Biology, 195(1), 27–40. https://doi.org/10.1083/jcb.201107093 1079 Onichtchouk, D., Glinka, A., & Niehrs, C. (1998). Requirement for Xvent-1 and Xvent-2 1080 gene function in dorsoventral patterning of Xenopus mesoderm. Development 1081 (Cambridge, England), 125(8), 1447–1456. 1082 Panagopoulos, I., Mencinger, M., Dietrich, C. U., Bjerkehagen, B., Saeter, G., Mertens, 1083 F., … Heim, S. (1999). Fusion of the RBP56 and CHN genes in extraskeletal 1084 myxoid chondrosarcomas with translocation t(9;17)(q22;q11). Oncogene, 18(52), 1085 7594–7598. https://doi.org/10.1038/sj.onc.1203155 1086 Pandya-Jones, A. (2011). Pre-mRNA splicing during transcription in the mammalian 1087 system. Wiley Interdisciplinary Reviews: RNA, 2(5), 700–717. 1088 https://doi.org/10.1002/wrna.86

Running Title: TAF15 regulates dorsoanterior neural development in Xenopus tropicalis Page 23

1089 Pandya-Jones, A., & Black, D. L. (2009). Co-transcriptional splicing of constitutive and 1090 alternative exons. RNA, 15(10), 1896–1908. https://doi.org/10.1261/rna.1714509 1091 Perales, R., & Bentley, D. (2009). “Cotranscriptionality”: The Transcription Elongation 1092 Complex as a Nexus for Nuclear Transactions. Molecular Cell, 36(2), 178–191. 1093 https://doi.org/10.1016/j.molcel.2009.09.018 1094 Rabbitts, T. H., Forster, A., Larson, R., & Nathan, P. (1993). Fusion of the dominant 1095 negative transcription regulator CHOP with a novel gene FUS by translocation 1096 t(12;16) in malignant liposarcoma. Nature Genetics, 4(2), 175–180. 1097 https://doi.org/10.1038/ng0693-175 1098 Rawat, V. P. S., Arseni, N., Ahmed, F., Mulaw, M. a, Thoene, S., Heilmeier, B., … 1099 Feuring-Buske, M. (2010). The vent-like homeobox gene VENTX promotes human 1100 myeloid differentiation and is highly expressed in acute myeloid leukemia. 1101 Proceedings of the National Academy of Sciences of the United States of America, 1102 107(39), 16946–16951. https://doi.org/10.1073/pnas.1001878107 1103 Reversade, B., Kurado, H., Lee, H. (2005). Depletion of Bmp2, Bmp4, Bmp7 and 1104 Spemann organizer signals induces massive brain formation in Xenopus embryos. 1105 Development, 132(15), 3381–3392. https://doi.org/10.1242/dev.01901 1106 Reyes, A., Anders, S., & Huber, W. (2014). Inferring differential exon usage in RNA-Seq 1107 data with the DEXSeq package. EMBL, Heidelberg, Germany. 1108 Sander, V., Reversade, B., & De Robertis, E. M. (2007). The opposing homeobox 1109 genes Goosecoid and Vent1/2 self-regulate Xenopus patterning. The EMBO 1110 Journal, 26(12), 2955–2965. https://doi.org/10.1038/sj.emboj.7601705 1111 Scekic‐Zahirovic, J., Sendscheid, O., El Oussini, H., Jambeau, M., Sun, Y., Mersmann, 1112 S., … Dupuis, L. (2016). Toxic gain of function from mutant FUS protein is crucial 1113 to trigger cell autonomous motor neuron loss . The EMBO Journal, 35(10), 1077– 1114 1097. https://doi.org/10.15252/embj.201592559 1115 Schuler-Metz, a, Knöchel, S., Kaufmann, E., & Knöchel, W. (2000). The homeodomain 1116 transcription factor Xvent-2 mediates autocatalytic regulation of BMP-4 expression 1117 in Xenopus embryos. The Journal of Biological Chemistry, 275(44), 34365–34374. 1118 https://doi.org/10.1074/jbc.M003915200 1119 Schwartz, J. C., Cech, T. R., & Parker, R. R. (2015). Biochemical Properties and 1120 Biological Functions of FET Proteins. Annual Review of Biochemistry, 84(1), 355– 1121 379. https://doi.org/10.1146/annurev-biochem-060614-034325 1122 Schwartz, J. C., Wang, X., Podell, E. R., & Cech, T. R. (2013). RNA Seeds Higher- 1123 Order Assembly of FUS Protein. Cell Reports, 5(4), 918–925. 1124 https://doi.org/10.1016/j.celrep.2013.11.017 1125 Sharma, A., Lyashchenko, A. K., Lu, L., Nasrabady, S. E., Elmaleh, M., Mendelsohn, 1126 M., … Shneider, N. A. (2016). ALS-associated mutant FUS induces selective motor 1127 neuron degeneration through toxic gain of function. Nature Communications, 7, 1– 1128 14. https://doi.org/10.1038/ncomms10465 1129 Shiohama, A., Sasaki, T., Noda, S., Minoshima, S., & Shimizu, N. (2007). Nucleolar 1130 localization of DGCR8 and identification of eleven DGCR8-associated proteins. 1131 Experimental Cell Research, 313(20), 4196–4207. 1132 https://doi.org/10.1016/j.yexcr.2007.07.020 1133 Sive, H.L., Grainger, R.M., Harland, R.M. 2000. Early development of Xenopus laevis. A 1134 laboratory manual. Cold Spring Harbor, New York: Cold Spring Harbor Laboratory 1135 Press. 1136 Sjögren, H., Meis-Kindblom, J., Kindblom, L. G., Åman, P., & Stenman, G. (1999).

Running Title: TAF15 regulates dorsoanterior neural development in Xenopus tropicalis Page 24

1137 Fusion of the EWS-related gene TAF2N to TEC in extraskeletal myxoid 1138 chondrosarcoma. Cancer Research, 59(20), 5064–5067. 1139 Svetoni, F., Frisone, P., & Paronetto, M. P. (2016). Role of FET proteins in 1140 neurodegenerative disorders. RNA Biology, 13(11), 1089–1102. 1141 https://doi.org/10.1080/15476286.2016.1211225 1142 Tan, A. Y., & Manley, J. L. (2009). The TET family of proteins: Functions and roles in 1143 disease. Journal of Molecular Cell Biology, 1(2), 82–92. 1144 https://doi.org/10.1093/jmcb/mjp025 1145 Trapnell, C., Roberts, A., Goff, L., Pertea, G., Kim, D., Kelley, D. R., … Pachter, L. 1146 (2012). Differential gene and transcript expression analysis of RNA-seq 1147 experiments with TopHat and Cufflinks. Nature Protocols, 7(3), 562–578. 1148 https://doi.org/10.1038/nprot.2012.016 1149 Vance, C., Rogelj, B., Hortobagyi, T., De Vos, K. J., Nishimura, A. L., Sreedharan, J., … 1150 Shaw, C. E. (2009). Mutations in FUS, an RNA Processing Protein, Cause Familial 1151 Amyotrophic Lateral Sclerosis Type 6. Science, 323(5918), 1208–1211. 1152 https://doi.org/10.1126/science.1165942 1153 Young, J. J., Kjolby, R. a S., Kong, N. R., Monica, S. D., & Harland, R. M. (2014). Spalt- 1154 like 4 promotes posterior neural fates via repression of pou5f3 family members in 1155 Xenopus. Development (Cambridge, England), 141(8), 1683–1693. 1156 https://doi.org/10.1242/dev.099374 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184

Running Title: TAF15 regulates dorsoanterior neural development in Xenopus tropicalis Page 25

1185 1186 FIGURE LEGENDS 1187 1188 Figure 1. taf15/TAF15 expression and depletion phenotype. (A) Representative 1189 whole mount in situ RNA hybridization to taf15 transcripts in X. tropicalis embryos. 1190 Stage 2-cell through stage 10; embryos were cut in half prior to hybridization. An = 1191 animal pole, Veg = vegetal pole, V = ventral, D = dorsal, Arrowhead = dorsal lip, R = 1192 right, L = left, A = anterior, P = posterior. (B) Western blot for TAF15 expression from 1193 two-cell through early tadpole; ZGA = zygotic genome activation. (C) Schematic 1194 indicating locations of morpholino and CRISPR/Cas9 target sites. (D) Translation- 1195 blocking (M+Z TAF15 depleted) and splice-blocking (Z TAF15 depleted) morpholino- 1196 mediated TAF15 depletion. A = anterior, P = posterior. (E) CRISPR/Cas9-mediated 1197 zygotic (Z)TAF15 depletion. A = anterior, P = posterior. (F) Representative Western blot 1198 for TAF15 protein expression in stage 15 embryos following morpholino or 1199 CRISPR/Cas9-mediated TAF15 depletion. Protein quantification is of the imaged blot 1200 using the most consistently expressed TAF15 band marked by an arrow, normalized to 1201 the corresponding ACTIN band; percent expression is relative to the uninjected 1202 condition; >3 blots were analyzed. (A) Representative images of >10 embryos. (D,E) 1203 Representative images of >12 embryos. (C,D,E) All injections for downstream imaging 1204 and Western blot analysis were into both cells of 2-cell stage embryos. 1205 1206 Figure 2. TAF15 depletion by translation-blocking morpholino leads to intron 1207 retention. (A) Schematic for maternal and zygotic TAF15 depletion via injection of 1208 translation-blocking morpholino and subsequent single embryo RNA-sequencing 1209 analysis. (B) PANTHER GO-Slim biological process classification of genes with 1210 differential exon usage following maternal and zygotic TAF15 depletion. (C) Genes with 1211 intron retention at both stages 10 and 15 classified by their maternal or zygotic 1212 expression, as well as the number of introns retained per transcript. **** P value = 1213 1.50723E-08. IR = intron retention. (D) qRT-PCR validation of RNA-seq results for 1214 genes with intron retention at both stages 10 and 15; qRT-PCR results expressed in fold 1215 change relative to uninjected controls; intron expression normalized to eef1a1 and to 1216 respective total gene expression; n = 3 individual embryos. n.s. = not significant. All 1217 injections for downstream RNA sequencing and qRT-PCR analysis were into both cells 1218 of 2-cell stage embryos. 1219 1220 Figure 3. TAF15 depletion by translation-blocking morpholino leads to single 1221 intron retention in fgfr4. (A) Visualization of fgfr4 RNA-seq reads with Integrative 1222 Genome Viewer aligned with gene model in blue. Red bars indicate qRT-PCR primers 1223 to measure retained intron. Blue bars indicated qRT-PCR primers to measure total 1224 transcript. UC = uninjected control; MO = M+Z TAF15-depleting morpholino (B) qRT- 1225 PCR for fgfr4 intron 1 expression in stage 15 embryos. Intron expression levels are 1226 normalized to total fgfr4 expression. *** P value = <0.0001. (C) qRT-PCR for total fgfr4 1227 expression on stage 15 embryos. ** P value = <0.005; *** P value = <0.001; **** P value 1228 = <0.0001. (B,C) Embryos treated as indicated on the x-axis; y-axis shows fgfr4 1229 expression relative to eef1a1 and normalized to uninjected embryos. Error bars = 1230 standard deviation; n = 9 individual embryos. All means were compared by one-way 1231 ANOVA followed by Tukey post-hoc analyses. All injections for downstream qRT-PCR 1232 analysis were into both cells of 2-cell stage embryos.

Running Title: TAF15 regulates dorsoanterior neural development in Xenopus tropicalis Page 26

1233 1234 Figure 4. TAF15 depletion leads to downregulation of FGFR4 and downstream 1235 targets of FGFR4. (A-E) Representative whole mount in situ hybridization to fgfr4 1236 transcripts in TAF15-depleted X. tropicalis embryos; * = injected side; D = dorsal; V = 1237 ventral; R = right; L = left. (F) Representative Western blot for TAF15 and FGFR4 1238 expression in stage 15 TAF15-depleted embryos. Protein quantification is of the imaged 1239 blot using the most consistently expressed TAF15 or FGFR4 band marked by an arrow, 1240 normalized to the corresponding ACTIN band; percent expression is relative to the 1241 uninjected condition; >3 blots were analyzed. (G-K) Representative whole mount in situ 1242 hybridization to pax2 transcripts in TAF15-depleted X. tropicalis embryos; n=9/condition; 1243 D = dorsal; V = ventral; R = right; L = left. (L) qRT-PCR for engrailed2 expression in 1244 stage 15 embryos treated as indicated on the x-axis; y-axis shows expression relative to 1245 eef1a1 and normalized to uninjected embryos; n = 6 individual embryos. * P value = 1246 <0.05; ** P value = 0.002; *** P value = <0.001; **** P value = <0.0001; Error bars = 1247 standard deviation. All means were compared by one-way ANOVA followed by Tukey 1248 post-hoc analyses. (A-E, G-K) Representative images of >12 embryos. All injections for 1249 downstream RNA ISH were into one cell of 2-cell stage embryos; uninjected side serves 1250 as internal control. (F,L) All injections for downstream qRT-PCR analysis were into both 1251 cells of 2-cell stage embryos. 1252 1253 Figure 5. TAF15 depletion leads to upregulation of the ventx family of 1254 transcription factors. (A) PANTHER GO-Slim Biological Process classification of 1255 differentially expressed genes following maternal and zygotic TAF15 depletion; stage 15 1256 X. tropicalis embryos. Asterisks mark Developmental Process genes combined into 1257 panel B. (B) Developmental Process subcategories. Boxed Ectoderm Development 1258 visualized in panel C. (C) Expression level heatmap of TAF15 target genes involved in 1259 Ectoderm development. Asterisks mark ventx family members. Columns represent 1260 RNAseq data from a single embryo. 1261 1262 Figure 6. taf15 and ventx2.1 exhibit complementary expression patterns and 1263 TAF15 depletion leads to expansion of ventx2.1 expression into the 1264 neurectoderm. (A,C,E,G,E’) Whole mount in situ hybridization to taf15. 1265 (B,D,F,H,F’,I,J,K) Whole mount in situ hybridization to ventx2.1. (A-D) Arrowhead = 1266 dorsal lip. (E’,F’) White dotted lines = boundaries between the neural ectoderm and 1267 lateral plate mesoderm; red dotted line = site of cross-section. (I,J) Arrowhead = eye 1268 enlage; white dotted line = site of cross-section. (K) white bracket = expanded ventx2.1 1269 expression. (L) Schematic comparing taf15 and ventx2.1 expression. Dark blue = 1270 stronger expression; light purple = weaker expression; red dotted line = region of 1271 ventx2.1 expansion. (M) Diagram of model for TAF15 repression of the ventrolateral 1272 BMP/Vent circuit. L = left; R = right; V = ventral; D = dorsal; A = anterior; P = posterior; s 1273 = sensorial layer of the epidermal ectoderm; e = epithelial layer of the epidermal 1274 ectoderm; n = neural ectoderm; lp = lateral plate mesoderm; * = injected side. (A-H) 1275 Representative images of >10 embryos. (E’,F’,K) Representative images of >5 cross- 1276 sectioned embryos. (I,J) Representative images of >12 embryos. (I,J,K) All injections for 1277 downstream RNA ISH were into one cell of 2-cell stage embryos; uninjected side serves 1278 as internal control. 1279

Running Title: TAF15 regulates dorsoanterior neural development in Xenopus tropicalis Page 27

1280 Figure 7. TAF15 depletion leads to increased and expanded ventx2.1 expression. 1281 (A) Visualization of ventx2.1 RNA-seq reads with Integrative Genome Viewer aligned 1282 with gene model in blue; UC = uninjected control; MO = M+Z TAF15-depleting 1283 morpholino. (B) qRT-PCR for ventx2.1 expression in stage 10 and 15 embryos treated 1284 as indicated on the x-axis; y-axis shows expression relative to eef1a1 and normalized to 1285 uninjected embryos; n = 9 individual embryos; * P value = <0.05; ** P value = <0.005; 1286 *** P value = <0.001; Error bars = standard deviation. All means were compared by 1287 one-way ANOVA followed by Tukey post-hoc analyses. All injections were into both 1288 cells of 2-cell stage embryos. (C-L) Representative whole mount in situ hybridization to 1289 ventx2.1; L = left; R = right; A = anterior; P = posterior; * = injected side. Representative 1290 images of >12 embryos. All injections for downstream RNA ISH were into one cell of 2- 1291 cell stage embryos; uninjected side serves as internal control. 1292 1293 Figure 8. Model of TAF15 pre- and post-transcriptional regulation. M TAF15 = 1294 maternal TAF15; M+Z TAF15 = maternal and zygotic TAF15; TBP = TATA-binding 1295 protein; RNA POL II = RNA polymerase 2. 1296 1297 1298 1299 1300

Running Title: TAF15 regulates dorsoanterior neural development in Xenopus tropicalis Page 28

A D E 2-cell Stage 6 Stage 8 Stage 9 Stage 10 Morpholino-mediated CRISPR/Cas9-mediated An TAF15 depletion TAF15 depletion

taf15 Vg V D

Stage 13 Stage 15 Uninjected Uninjected Anterior Dorsal Anterior Dorsal P Z TAF15 depleted depleted M+Z TAF15 M+Z

R L A Stage 26 Stage 32 Z TAF15 depleted Cas9 only Cas9

A P

B Mismatch

ZGA Morpholino A P

kDA Ladder 2-cell 8-cell 32-cell 8 Stage 9 Stage 10 Stage 10.5/11 Stage 12/12.5 Stage 15 Stage 20 Stage 22 Stage 26 Stage 30 Stage 75 α-TAF15 50 Ladder Uninjected M+Z TAF15 depleted Z TAF15 depleted kDa Mismatch MO Cas9 only Cas9 kDa Ladder Uninjected taf15 sgRNA + Cas9 50 75 75 α-ACTIN α-TAF15 α-TAF15

37 50 50

~42 α-ACTIN ~42 α-ACTIN C Translation-blocking Splice-blocking M+Z taf15 MO Z taf15 MO % TAF15 expression: 100 30 46 100 100 76 100

Exon 1 ATG Exon 2 Intron 2 ... Exon 16

taf15 sgRNA CRISPR Cas9 DeJong et al. Figure 1 bioRxiv preprint doi: https://doi.org/10.1101/2021.06.14.041913; this version posted June 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

DEXseq A exon usage Illumina TruSeq Kit TopHat HTseq-count UC : MO Make single embryo Align reads to Count transcript DESeq RNA-seq library X.t. genome v.7.1 abundance M+Z TAF15 depleted embryo transcript expression UC : MO B Stage 10 Stage 15 PANTHER GO-Slim Biological Process PANTHER GO-Slim Biological Process PANTHER GO-Slim Biological Process Total # Genes: 146 Total # Process hits: 264 Total # Genes: 788 Total # Process hits: 1329 Total # Genes: 86 Total # Process hits: 142 85 400 80 45 75 70 350 40 65 60 300 35 55 50 250 30 45 25 40 200 35 20 # of Genes # of Genes 30 150 # of Genes 25 15 20 100 15 10 50 10 5 5 0 0 0 Localization (GO: 0051179) (GO: Localization Localization (GO: 0051179) (GO: Localization Localization (GO: 0051179) (GO: Localization Locomotion (GO: 0040011) (GO: Locomotion Locomotion (GO: 0040011) (GO: Locomotion Locomotion (GO: 0040011) (GO: Locomotion Reproduction (GO: 0000003) Reproduction (GO: Reproduction (GO: 0000003) Reproduction (GO: Reproduction (GO: 0000003) Reproduction (GO: Cellular Process (GO: 0009987) (GO: Process Cellular Cellular Process (GO: 0009987) (GO: Process Cellular Cellular Process (GO: 0009987) (GO: Process Cellular Metabolic Process (GO: 0008152) (GO: Metabolic Process Metabolic Process (GO: 0008152) (GO: Metabolic Process Metabolic Process (GO: 0008152) (GO: Metabolic Process Apoptotic Process (GO: 0006915) (GO: Process Apoptotic Apoptotic Process (GO: 0006915) (GO: Process Apoptotic Apoptotic Process (GO: 0006915) (GO: Process Apoptotic Biological Adhesion (GO: 0022610) (GO: Adhesion Biological Biological Adhesion (GO: 0022610) (GO: Adhesion Biological Biological Adhesion (GO: 0022610) (GO: Adhesion Biological Biological Regulation (GO: 0065007) Regulation (GO: Biological Biological Regulation (GO: 0065007) Regulation (GO: Biological Biological Regulation (GO: 0065007) Regulation (GO: Biological Response to Stimulus (GO: 0050896) (GO: Stimulus Response to Response to Stimulus (GO: 0050896) (GO: Stimulus Response to Response to Stimulus (GO: 0050896) (GO: Stimulus Response to Developmental Process (GO: 0032502) (GO: Process Developmental Developmental Process (GO: 0032502) (GO: Process Developmental Developmental Process (GO: 0032502) (GO: Process Developmental Immune System Process (GO: 0002376) (GO: Process Immune System Immune System Process (GO: 0002376) (GO: Process Immune System Immune System Process (GO: 0002376) (GO: Process Immune System Multicellular Organismal Process (GO: 0032501) (GO: Process Organismal Multicellular Multicellular Organismal Process (GO: 0032501) (GO: Process Organismal Multicellular Multicellular Organismal Process (GO: 0032501) (GO: Process Organismal Multicellular Cellular Component Organization or Biogenesis (GO: 0071840) or Biogenesis (GO: Organization Component Cellular Cellular Component Organization or Biogenesis (GO: 0071840) or Biogenesis (GO: Organization Component Cellular Cellular Component Organization or Biogenesis (GO: 0071840) or Biogenesis (GO: Organization Component Cellular

C Stage-persistent intron retention D Validation of stage-persistent intron retentions

**** RNA-Seq qRT-PCR 100 Morpholino

80 Stage 10 Stage 15 Stage 15 Intron Maternally M+Z TAF15 M+Z TAF15 Z TAF15 60 Gene name retained expressed? Depleted Depleted Depleted Mismatch G ene s 40 o f bmp7.1 3 yes 5.80 3.08 2.29 n.s. n.s.

% 20 clp1 within 5’ UTR yes <2.0 2.33 n.s. n.s. n.s. dgka 11 yes 5.55 5.64 4.04 n.s. n.s. 0 dgka 9 yes 2.09 3.44 2.14 n.s. n.s. Observed Expected fgfr4 1 yes 3.11 2.96 5.9 n.s. n.s. Maternal 92 65 pdgfa 1 yes <2.0 2.83 n.s. n.s. n.s. Zygotic 8 35 per2 1 yes 2.95 4.47 5.66 n.s. n.s. rab15 part of 2 yes 2.07 2.75 3.93 n.s. n.s. 100 srsf4 5 yes <2.0 3.51 5.7 n.s. n.s. zdhhc5 part of 6 yes <2.0 3.16 2.10 n.s. n.s. 80 gpr110 9 no 5.18 23.41 n.s. n.s. n.s. 60 isl1 3 no 3.13 3.08 5.2 n.s. n.s.

Genes 40

# o f 20

0 # of retained introns/transcript 1 IR 83 2 IR 13 DeJong et al. Figure 2 >2 IR 4 bioRxiv preprint doi: https://doi.org/10.1101/2021.06.14.041913; this version posted June 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

A fgfr4

2030

UC 0 2030 Stage 10 Stage MO 0

2425

UC 0 2425 Stage 15 Stage MO 0

8 8 *** UC st.15 *** UC st.15 B fgfr4fgfr4 intron1 intron expression/total fgfr4 expressionfgfr4 intron1 Normalized fgfr4 intrMon + Z expreDepletionssion NorNormmaliToalitalzez dedfgfr4 fgf fgfr 4er4xpr int intrerMssonon +ion Z exp expreDepletionressiossionn 6 fgfr4 c-terminus 6 Z*** Depletion Z*** Depletion *** *** 8 *** Mismatch MO 2.01.58 *** Mismatch MO 4 UC st.15 ** UC4UC st.1 s M + Z Depletion M M+ Z+ D 6 1.5 6 2 Z Depletion1 Z2 DeplZ De Mismatch MO expression Relative MismaMism ss ion 4 ss io n 1.0 4 d Nor malize d 0 Nor malize d 0 xpr e e to Uninjected e to Uninjected .5 E 2 Expre 0.5 2 tch MO ch MO Relative expression expression Relative Relative expression expression Relative Relative expression expression Relative UC st.15 Depletion expression Relative Depletion UC st.15 Depletion Depletion UC st.15 Z Z Relati ve + Z Relati ve 100% 136% 69% 107% + Z Relativ Mismat

Misma Relativ M 0 M 0.0 00 d d d O cted h M h MO eplete epleted epleted d deplete ninje Cas9 only Uninjecte UninUjected ismatc ismatc M sgRNA + Cas9 M TAF15 TAF15 fgfr4taf15 Dc-terminustaf15 D Z Z 8 8 M+Z taf15 M+Z ***Myriad Pro UC st.15 UC st.15Myriad Pro C ** = p < 0.005 fgfr4 intron1 intron expressionetion *** = p < 0.0001Total fgfr4 expression Total fgfr4 expression*** =Norm p < 0.0001aliTotalzed fgf fgfr4r4 intr eMxpreon + Z expreDepletionssionssionM + Z Depleti fgfr4 c-terminus 6 6 n Z*** Depletion Z Depletion Error bars = SD ErrorError bars bars = SD= SD *** MO 2 81.5 *** Mismatch MO Mismatch MO UC st.15 4 UC4 st.15 *** UC s ** **** M + Z Depletion n.s. M + Z Depletion M + 1.5 6 ** Z Depletion 2 Z2 Depletion1 Z De

Relative expression expression Relative Mism Mismatch MO Mismatch MO ss ion 1 4

0 0 Nor malize d

e to Uninjected .5 Ex pr e

.5 ti v 2 ch MO Relative expression expression Relative Relative expression expression Relative Relative expression expression Relative UC st.15 Depletion UC st.Deple15 tion Depletion Depletion Z Z Relati ve

+ Z Rela + Z Mismat

Relative to Uninjected Relative Mismatch MO 0 M 0M0 ed ed ed cted ch MO h MO epleted epleted deplet deplet ninje Cas9 only Uninject UninjeUcted Mismat sgRNA + Cas9 Mismatc TAF15 TAF15 taf15 D taf15 D Z Z taf15 M+Z M+Z Myriad Pro ** = p < 0.005 *** = p < 0.0001

Error bars = SD Error bars = SD

**** p= <0.0001 DeJong et al. Figure*** p= <0.0013 bioRxiv preprint doi: https://doi.org/10.1101/2021.06.14.041913; this version posted June 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

M+Z TAF15 depleted Z TAF15 depleted Mismatch MO taf15 sgRNA+ Cas9 Cas9 only A D B C D E Anterior

fgfr4 * V R * L * * * F Uninjected depleted M+Z TAF15 depleted Z TAF15 kDa Ladder Uninjected depleted M+Z TAF15 depleted Z TAF15 kDa Ladder 75 75 α-FGFR4 α-TAF15 50 50 100 35 52 % Expressed 100 20 21 % Expressed 50 50 α-ACTIN -ACTIN α 37 37

M+Z TAF15 depleted Z TAF15 depleted Mismatch MO taf15 sgRNA+ Cas9 Cas9 only G D H I J K Anterior

pax2 * V R * L * * * L engrailed 2 engrailed 2 ** 1.5 * 1.5 *** ****

1.0 1.0

0.5 0.5 Relative to Uninjected Relative to Uninjected 0.0 0.0

as9 only jected eted C 9 ch MO njected + Cas Unin depl mat Uni NA Mis sgR TAF15 Z TAF15 depleted +Z taf15 M DeJong et al. Figure 4 bioRxiv preprint doi: https://doi.org/10.1101/2021.06.14.041913; this version posted June 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

A B Decreased Expression Increased Expression PANTHER GO-Slim Biological Process PANTHER GO-Slim Biological Process Developmental Process (GO: 0032502) Total # Genes: 773 Total # Process hits: 1485 Total # Genes: 462 Total # Process hits: 789 Total # Genes: 207 Total # Process hits: 390 130 360 260 340 240 120 320 220 110 300 100 280 200 260 180 90 240 80 220 160 200 140 70 180 60 160 120 # of Genes # of Genes # of Genes 140 * 100 50 120 80 40 100 80 60 * 30 60 40 20 40 20 20 10 0 0 0 Death (GO: 0016265) Death (GO: Growth (GO: 0040007) (GO: Growth Growth (GO: 0040007) (GO: Growth Localization (GO: 0051179) (GO: Localization Localization (GO: 0051179) (GO: Localization Locomotion (GO: 0040011) (GO: Locomotion Locomotion (GO: 0040011) (GO: Locomotion Reproduction (GO: 0000003) Reproduction (GO: Reproduction (GO: 0000003) Reproduction (GO: Cellular Process (GO: 0009987) (GO: Process Cellular Cellular Process (GO: 0009987) (GO: Process Cellular Metabolic Process (GO: 0008152) (GO: Metabolic Process Metabolic Process (GO: 0008152) (GO: Metabolic Process Apoptotic Process (GO: 0006915) (GO: Process Apoptotic Apoptotic Process (GO: 0006915) (GO: Process Apoptotic Sex Determination (GO: 0007530) (GO: Sex Determination Biological Adhesion (GO: 0022610) (GO: Adhesion Biological Biological Adhesion (GO: 0022610) (GO: Adhesion Biological Biological Regulation (GO: 0065007) (GO: Regulation Biological Biological Regulation (GO: 0065007) (GO: Regulation Biological Response to Stimulus (GO: 0050896) (GO: Stimulus Response to Response to Stimulus (GO: 0050896) (GO: Stimulus Response to System Development (GO: 0048731) (GO: Development System Embryo Development (GO: 0009790) Embryo (GO: Development Developmental Process (GO: 0032502) (GO: Process Developmental Developmental Process (GO: 0032502) (GO: Process Developmental Immune System Process (GO: 0002376) (GO: Process Immune System Immune System Process (GO: 0002376) (GO: Process Immune System Ectoderm Development (GO: 0007398) (GO: Ectoderm Development Mesoderm Development (GO: 0007498) Mesoderm (GO: Development Multicellular Organismal Process (GO: 0032501) (GO: Process Organismal Multicellular Multicellular Organismal Process (GO: 0032501) (GO: Process Organismal Multicellular Anatomical Structure Morphogenesis (GO: 0009653) Structure Morphogenesis (GO: Anatomical Cellular Component Organization/Biogenesis (GO: 0071840) (GO: Organization/Biogenesis Component Cellular Cellular Component Organization/Biogenesis (GO: 0071840) (GO: Organization/Biogenesis Component Cellular C −1.5 0 1 Distance

en2 rbm23 hoxb2 hoxc3 mtss1l dbx1 phox2a olig3 ntng2 sema3b tlx1 dct pax1 thsd7b pax7 atrnl1 f5 tnfrsf1a emp1 fabp7 igsf10 slitrk2 barx1 fas atoh1 lhx9 lamb2 phox2b ventx3.1 * cdk14 mixer ventx2.2 * isl1 ventx2.1 * hoxd4 ifrd1 odz4 tll2 lmo4.2 bcam btbd17 lhx5 evx1 ventx1.1 * DeJong et al. dscaml1 Figure 5 Uninjected M+Z TAF15 depleted bioRxiv preprint doi: https://doi.org/10.1101/2021.06.14.041913; this version posted June 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Dorsal Dorsoventral cross section taf15 ventx2.1 taf15 ventx2.1 A B C D Stage 10 Stage

L R V D Anterior E P F G D H Stage 15 Stage

A V

Dorsal cross section E’ F’ n D n

Dorsal lp e P s

taf15 ventx2.1 V A Anterior Dorsal Cross section I J P K n

R L A ventx2.1 * * ventx2.1 * L Dorsal M Anterior P Dorsal TAF15

R L taf15 ventx2.1ventx2.1 A Ventral BMP VENT DeJong et al. Figure 6 bioRxiv preprint doi: https://doi.org/10.1101/2021.06.14.041913; this version posted June 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

A ventx2.1

9025

UC 0 9025 Stage 10 Stage MO 0 7320

UC 0 7320 Stage 15 Stage MO 0

B ventx2.1 Stage 10 Stage 15 st10 ventx2.1 ventx2.1 3 3 3 ** UC st.10 * UC st.15 ** M + Z Depletion M + Z Depletion *** * * *** 2 2Z Depletion Z 2Depletion Mismatch MO Mismatch MO

1 1 1 Relative expression expression Relative Relative expression expression Relative Relative to Uninjected 0 Relative to Uninjected 0 Relative to Uninjected 0

injected depleted depleted n Cas9 only Uninjected depleted depleted Uninjected U Mismatch MO Mismatch MO sgRNA + Cas9 Z TAF15 Z TAF15 taf15 M+Z TAF15 M+Z TAF15 * * * = p < 0.05 * * * Error bars = SD M+Z TAF15 Depleted Z TAF15 Depleted Mismatch MO taf15 sgRNA + Cas9 Cas9 only < 0.05 < 0.0011C D E *** p= F<0.001 G bars = SD Anterior R L ventx2.1 * * * ** ** H P I J K L Dorsal

A V * * * * *

DeJong et al. Figure 7 bioRxiv preprint doi: https://doi.org/10.1101/2021.06.14.041913; this version posted June 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Post-transcriptional regulation Transcriptional regulation

Maternal and zygotic transcripts Zygotic transcripts M+Z M Splice TAF15 RNA TAF15 factor POL II TBP Target gene exon intron Core promoter

DeJong et al. Figure 8